Upload 132 files
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- qnli/bert-base-uncased_lr1e-05/config.json +26 -0
- qnli/bert-base-uncased_lr1e-05/model.safetensors +3 -0
- qnli/bert-base-uncased_lr1e-05/special_tokens_map.json +7 -0
- qnli/bert-base-uncased_lr1e-05/tokenizer.json +0 -0
- qnli/bert-base-uncased_lr1e-05/tokenizer_config.json +55 -0
- qnli/bert-base-uncased_lr1e-05/trainer_state.json +951 -0
- qnli/bert-base-uncased_lr1e-05/training_args.bin +3 -0
- qnli/bert-base-uncased_lr1e-05/vocab.txt +0 -0
- qnli/roberta-base_lr1e-05/classifier_head.pt +3 -0
- qnli/roberta-base_lr1e-05/config.json +27 -0
- qnli/roberta-base_lr1e-05/merges.txt +0 -0
- qnli/roberta-base_lr1e-05/model.safetensors +3 -0
- qnli/roberta-base_lr1e-05/special_tokens_map.json +15 -0
- qnli/roberta-base_lr1e-05/tokenizer.json +0 -0
- qnli/roberta-base_lr1e-05/tokenizer_config.json +57 -0
- qnli/roberta-base_lr1e-05/trainer_state.json +951 -0
- qnli/roberta-base_lr1e-05/training_args.bin +3 -0
- qnli/roberta-base_lr1e-05/vocab.json +0 -0
- qnli/roberta-large_lr1e-05/classifier_head.pt +3 -0
- qnli/roberta-large_lr1e-05/config.json +27 -0
- qnli/roberta-large_lr1e-05/merges.txt +0 -0
- qnli/roberta-large_lr1e-05/model.safetensors +3 -0
- qnli/roberta-large_lr1e-05/special_tokens_map.json +15 -0
- qnli/roberta-large_lr1e-05/tokenizer.json +0 -0
- qnli/roberta-large_lr1e-05/tokenizer_config.json +57 -0
- qnli/roberta-large_lr1e-05/trainer_state.json +951 -0
- qnli/roberta-large_lr1e-05/training_args.bin +3 -0
- qnli/roberta-large_lr1e-05/vocab.json +0 -0
- qqp/bert-base-uncased_lr1e-05/config.json +26 -0
- qqp/bert-base-uncased_lr1e-05/model.safetensors +3 -0
- qqp/bert-base-uncased_lr1e-05/special_tokens_map.json +7 -0
- qqp/bert-base-uncased_lr1e-05/tokenizer.json +0 -0
- qqp/bert-base-uncased_lr1e-05/tokenizer_config.json +55 -0
- qqp/bert-base-uncased_lr1e-05/trainer_state.json +3015 -0
- qqp/bert-base-uncased_lr1e-05/training_args.bin +3 -0
- qqp/bert-base-uncased_lr1e-05/vocab.txt +0 -0
- qqp/roberta-base_lr1e-05/classifier_head.pt +3 -0
- qqp/roberta-base_lr1e-05/config.json +27 -0
- qqp/roberta-base_lr1e-05/merges.txt +0 -0
- qqp/roberta-base_lr1e-05/model.safetensors +3 -0
- qqp/roberta-base_lr1e-05/special_tokens_map.json +15 -0
- qqp/roberta-base_lr1e-05/tokenizer.json +0 -0
- qqp/roberta-base_lr1e-05/tokenizer_config.json +57 -0
- qqp/roberta-base_lr1e-05/trainer_state.json +3015 -0
- qqp/roberta-base_lr1e-05/training_args.bin +3 -0
- qqp/roberta-base_lr1e-05/vocab.json +0 -0
- qqp/roberta-large_lr1e-05/classifier_head.pt +3 -0
- qqp/roberta-large_lr1e-05/config.json +27 -0
- qqp/roberta-large_lr1e-05/merges.txt +0 -0
- qqp/roberta-large_lr1e-05/model.safetensors +3 -0
qnli/bert-base-uncased_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/shikexuan/nlu_model/bert-base-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.45.2",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
qnli/bert-base-uncased_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e11c03e7a720033667c8f3f7d6014a075b5720f09c0070601b14f102f90a7f5
|
| 3 |
+
size 437958648
|
qnli/bert-base-uncased_lr1e-05/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
qnli/bert-base-uncased_lr1e-05/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/bert-base-uncased_lr1e-05/tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
qnli/bert-base-uncased_lr1e-05/trainer_state.json
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8998568019093078,
|
| 3 |
+
"best_model_checkpoint": "./nlu_finetuned_models/qnli/bert-base-uncased_lr1e-05/checkpoint-58920",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 58920,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08486082824168364,
|
| 13 |
+
"grad_norm": 5.244738578796387,
|
| 14 |
+
"learning_rate": 1.4140271493212672e-06,
|
| 15 |
+
"loss": 0.7656,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1697216564833673,
|
| 20 |
+
"grad_norm": 7.798401355743408,
|
| 21 |
+
"learning_rate": 2.8280542986425343e-06,
|
| 22 |
+
"loss": 0.5938,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.2545824847250509,
|
| 27 |
+
"grad_norm": 6.452247619628906,
|
| 28 |
+
"learning_rate": 4.242081447963801e-06,
|
| 29 |
+
"loss": 0.4848,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3394433129667346,
|
| 34 |
+
"grad_norm": 14.68191909790039,
|
| 35 |
+
"learning_rate": 5.656108597285069e-06,
|
| 36 |
+
"loss": 0.4622,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4243041412084182,
|
| 41 |
+
"grad_norm": 6.4854583740234375,
|
| 42 |
+
"learning_rate": 7.070135746606336e-06,
|
| 43 |
+
"loss": 0.4523,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5091649694501018,
|
| 48 |
+
"grad_norm": 7.5430426597595215,
|
| 49 |
+
"learning_rate": 8.484162895927603e-06,
|
| 50 |
+
"loss": 0.4328,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.5940257976917854,
|
| 55 |
+
"grad_norm": 10.789835929870605,
|
| 56 |
+
"learning_rate": 9.89819004524887e-06,
|
| 57 |
+
"loss": 0.4091,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6788866259334692,
|
| 62 |
+
"grad_norm": 7.9694647789001465,
|
| 63 |
+
"learning_rate": 9.916221291347683e-06,
|
| 64 |
+
"loss": 0.3998,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.7637474541751528,
|
| 69 |
+
"grad_norm": 6.613808631896973,
|
| 70 |
+
"learning_rate": 9.825942510472339e-06,
|
| 71 |
+
"loss": 0.3657,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.8486082824168364,
|
| 76 |
+
"grad_norm": 6.988012790679932,
|
| 77 |
+
"learning_rate": 9.735663729596995e-06,
|
| 78 |
+
"loss": 0.3446,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.93346911065852,
|
| 83 |
+
"grad_norm": 5.471568584442139,
|
| 84 |
+
"learning_rate": 9.645384948721653e-06,
|
| 85 |
+
"loss": 0.3358,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.0,
|
| 90 |
+
"eval_accuracy": 0.874653937947494,
|
| 91 |
+
"eval_loss": 0.3010239005088806,
|
| 92 |
+
"eval_runtime": 7.3193,
|
| 93 |
+
"eval_samples_per_second": 1431.154,
|
| 94 |
+
"eval_steps_per_second": 89.49,
|
| 95 |
+
"step": 5892
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.0183299389002036,
|
| 99 |
+
"grad_norm": 12.543919563293457,
|
| 100 |
+
"learning_rate": 9.55510616784631e-06,
|
| 101 |
+
"loss": 0.3139,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1031907671418872,
|
| 106 |
+
"grad_norm": 13.244446754455566,
|
| 107 |
+
"learning_rate": 9.464827386970968e-06,
|
| 108 |
+
"loss": 0.2878,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1880515953835709,
|
| 113 |
+
"grad_norm": 1.526010274887085,
|
| 114 |
+
"learning_rate": 9.374548606095624e-06,
|
| 115 |
+
"loss": 0.2763,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.2729124236252547,
|
| 120 |
+
"grad_norm": 12.79552173614502,
|
| 121 |
+
"learning_rate": 9.28426982522028e-06,
|
| 122 |
+
"loss": 0.2863,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.3577732518669383,
|
| 127 |
+
"grad_norm": 8.889263153076172,
|
| 128 |
+
"learning_rate": 9.193991044344937e-06,
|
| 129 |
+
"loss": 0.2718,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.442634080108622,
|
| 134 |
+
"grad_norm": 13.187612533569336,
|
| 135 |
+
"learning_rate": 9.103712263469595e-06,
|
| 136 |
+
"loss": 0.2768,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.5274949083503055,
|
| 141 |
+
"grad_norm": 8.599838256835938,
|
| 142 |
+
"learning_rate": 9.013433482594251e-06,
|
| 143 |
+
"loss": 0.2638,
|
| 144 |
+
"step": 9000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.6123557365919892,
|
| 148 |
+
"grad_norm": 8.132196426391602,
|
| 149 |
+
"learning_rate": 8.92315470171891e-06,
|
| 150 |
+
"loss": 0.2754,
|
| 151 |
+
"step": 9500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.6972165648336728,
|
| 155 |
+
"grad_norm": 1.9769632816314697,
|
| 156 |
+
"learning_rate": 8.832875920843566e-06,
|
| 157 |
+
"loss": 0.2623,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.7820773930753564,
|
| 162 |
+
"grad_norm": 6.259141445159912,
|
| 163 |
+
"learning_rate": 8.742597139968224e-06,
|
| 164 |
+
"loss": 0.2735,
|
| 165 |
+
"step": 10500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.86693822131704,
|
| 169 |
+
"grad_norm": 12.646600723266602,
|
| 170 |
+
"learning_rate": 8.65231835909288e-06,
|
| 171 |
+
"loss": 0.257,
|
| 172 |
+
"step": 11000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.9517990495587236,
|
| 176 |
+
"grad_norm": 5.906457424163818,
|
| 177 |
+
"learning_rate": 8.562039578217536e-06,
|
| 178 |
+
"loss": 0.258,
|
| 179 |
+
"step": 11500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_accuracy": 0.8979474940334129,
|
| 184 |
+
"eval_loss": 0.27632975578308105,
|
| 185 |
+
"eval_runtime": 7.3451,
|
| 186 |
+
"eval_samples_per_second": 1426.128,
|
| 187 |
+
"eval_steps_per_second": 89.176,
|
| 188 |
+
"step": 11784
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.0366598778004072,
|
| 192 |
+
"grad_norm": 16.377243041992188,
|
| 193 |
+
"learning_rate": 8.471760797342193e-06,
|
| 194 |
+
"loss": 0.2263,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.121520706042091,
|
| 199 |
+
"grad_norm": 19.77906036376953,
|
| 200 |
+
"learning_rate": 8.381482016466849e-06,
|
| 201 |
+
"loss": 0.1759,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.2063815342837745,
|
| 206 |
+
"grad_norm": 4.166837215423584,
|
| 207 |
+
"learning_rate": 8.291203235591507e-06,
|
| 208 |
+
"loss": 0.1801,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.291242362525458,
|
| 213 |
+
"grad_norm": 2.036400318145752,
|
| 214 |
+
"learning_rate": 8.200924454716163e-06,
|
| 215 |
+
"loss": 0.1899,
|
| 216 |
+
"step": 13500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.3761031907671417,
|
| 220 |
+
"grad_norm": 9.122848510742188,
|
| 221 |
+
"learning_rate": 8.110645673840822e-06,
|
| 222 |
+
"loss": 0.1889,
|
| 223 |
+
"step": 14000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.4609640190088253,
|
| 227 |
+
"grad_norm": 14.528931617736816,
|
| 228 |
+
"learning_rate": 8.020366892965478e-06,
|
| 229 |
+
"loss": 0.1887,
|
| 230 |
+
"step": 14500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.5458248472505094,
|
| 234 |
+
"grad_norm": 9.965221405029297,
|
| 235 |
+
"learning_rate": 7.930088112090136e-06,
|
| 236 |
+
"loss": 0.1931,
|
| 237 |
+
"step": 15000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.6306856754921926,
|
| 241 |
+
"grad_norm": 25.76782989501953,
|
| 242 |
+
"learning_rate": 7.839809331214792e-06,
|
| 243 |
+
"loss": 0.196,
|
| 244 |
+
"step": 15500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.7155465037338766,
|
| 248 |
+
"grad_norm": 6.8294782638549805,
|
| 249 |
+
"learning_rate": 7.749530550339449e-06,
|
| 250 |
+
"loss": 0.1974,
|
| 251 |
+
"step": 16000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.8004073319755602,
|
| 255 |
+
"grad_norm": 2.1988823413848877,
|
| 256 |
+
"learning_rate": 7.659251769464105e-06,
|
| 257 |
+
"loss": 0.1874,
|
| 258 |
+
"step": 16500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.885268160217244,
|
| 262 |
+
"grad_norm": 15.78819465637207,
|
| 263 |
+
"learning_rate": 7.568972988588763e-06,
|
| 264 |
+
"loss": 0.1863,
|
| 265 |
+
"step": 17000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.9701289884589275,
|
| 269 |
+
"grad_norm": 20.404821395874023,
|
| 270 |
+
"learning_rate": 7.478694207713419e-06,
|
| 271 |
+
"loss": 0.2015,
|
| 272 |
+
"step": 17500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.0,
|
| 276 |
+
"eval_accuracy": 0.8986157517899761,
|
| 277 |
+
"eval_loss": 0.31375744938850403,
|
| 278 |
+
"eval_runtime": 7.325,
|
| 279 |
+
"eval_samples_per_second": 1430.037,
|
| 280 |
+
"eval_steps_per_second": 89.42,
|
| 281 |
+
"step": 17676
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.054989816700611,
|
| 285 |
+
"grad_norm": 1.0869024991989136,
|
| 286 |
+
"learning_rate": 7.388415426838077e-06,
|
| 287 |
+
"loss": 0.1529,
|
| 288 |
+
"step": 18000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.1398506449422947,
|
| 292 |
+
"grad_norm": 94.14501190185547,
|
| 293 |
+
"learning_rate": 7.298136645962734e-06,
|
| 294 |
+
"loss": 0.1117,
|
| 295 |
+
"step": 18500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.2247114731839783,
|
| 299 |
+
"grad_norm": 5.619855880737305,
|
| 300 |
+
"learning_rate": 7.207857865087391e-06,
|
| 301 |
+
"loss": 0.135,
|
| 302 |
+
"step": 19000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.309572301425662,
|
| 306 |
+
"grad_norm": 9.343029975891113,
|
| 307 |
+
"learning_rate": 7.117579084212047e-06,
|
| 308 |
+
"loss": 0.1306,
|
| 309 |
+
"step": 19500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.3944331296673456,
|
| 313 |
+
"grad_norm": 23.610340118408203,
|
| 314 |
+
"learning_rate": 7.027300303336704e-06,
|
| 315 |
+
"loss": 0.135,
|
| 316 |
+
"step": 20000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.479293957909029,
|
| 320 |
+
"grad_norm": 47.40991973876953,
|
| 321 |
+
"learning_rate": 6.937021522461362e-06,
|
| 322 |
+
"loss": 0.1259,
|
| 323 |
+
"step": 20500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.564154786150713,
|
| 327 |
+
"grad_norm": 6.907505512237549,
|
| 328 |
+
"learning_rate": 6.846742741586018e-06,
|
| 329 |
+
"loss": 0.1493,
|
| 330 |
+
"step": 21000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.6490156143923964,
|
| 334 |
+
"grad_norm": 0.08741018921136856,
|
| 335 |
+
"learning_rate": 6.756463960710675e-06,
|
| 336 |
+
"loss": 0.1329,
|
| 337 |
+
"step": 21500
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.73387644263408,
|
| 341 |
+
"grad_norm": 37.258567810058594,
|
| 342 |
+
"learning_rate": 6.6661851798353315e-06,
|
| 343 |
+
"loss": 0.1508,
|
| 344 |
+
"step": 22000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.8187372708757636,
|
| 348 |
+
"grad_norm": 0.5979816913604736,
|
| 349 |
+
"learning_rate": 6.57590639895999e-06,
|
| 350 |
+
"loss": 0.1484,
|
| 351 |
+
"step": 22500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.9035980991174473,
|
| 355 |
+
"grad_norm": 12.651761054992676,
|
| 356 |
+
"learning_rate": 6.485627618084646e-06,
|
| 357 |
+
"loss": 0.1492,
|
| 358 |
+
"step": 23000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.988458927359131,
|
| 362 |
+
"grad_norm": 13.937551498413086,
|
| 363 |
+
"learning_rate": 6.395348837209303e-06,
|
| 364 |
+
"loss": 0.1279,
|
| 365 |
+
"step": 23500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.0,
|
| 369 |
+
"eval_accuracy": 0.8991885441527446,
|
| 370 |
+
"eval_loss": 0.4502042233943939,
|
| 371 |
+
"eval_runtime": 7.3149,
|
| 372 |
+
"eval_samples_per_second": 1432.002,
|
| 373 |
+
"eval_steps_per_second": 89.543,
|
| 374 |
+
"step": 23568
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.0733197556008145,
|
| 378 |
+
"grad_norm": 40.33448791503906,
|
| 379 |
+
"learning_rate": 6.3050700563339595e-06,
|
| 380 |
+
"loss": 0.0896,
|
| 381 |
+
"step": 24000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.158180583842499,
|
| 385 |
+
"grad_norm": 2.8410744667053223,
|
| 386 |
+
"learning_rate": 6.214791275458617e-06,
|
| 387 |
+
"loss": 0.0998,
|
| 388 |
+
"step": 24500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.243041412084182,
|
| 392 |
+
"grad_norm": 29.884593963623047,
|
| 393 |
+
"learning_rate": 6.124512494583274e-06,
|
| 394 |
+
"loss": 0.1086,
|
| 395 |
+
"step": 25000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.327902240325866,
|
| 399 |
+
"grad_norm": 47.791419982910156,
|
| 400 |
+
"learning_rate": 6.034233713707931e-06,
|
| 401 |
+
"loss": 0.0825,
|
| 402 |
+
"step": 25500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.412763068567549,
|
| 406 |
+
"grad_norm": 0.26600632071495056,
|
| 407 |
+
"learning_rate": 5.943954932832587e-06,
|
| 408 |
+
"loss": 0.1024,
|
| 409 |
+
"step": 26000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.497623896809233,
|
| 413 |
+
"grad_norm": 16.983444213867188,
|
| 414 |
+
"learning_rate": 5.853676151957245e-06,
|
| 415 |
+
"loss": 0.1098,
|
| 416 |
+
"step": 26500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.582484725050916,
|
| 420 |
+
"grad_norm": 11.912595748901367,
|
| 421 |
+
"learning_rate": 5.763397371081901e-06,
|
| 422 |
+
"loss": 0.0967,
|
| 423 |
+
"step": 27000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.6673455532926,
|
| 427 |
+
"grad_norm": 0.07782808691263199,
|
| 428 |
+
"learning_rate": 5.673118590206559e-06,
|
| 429 |
+
"loss": 0.1183,
|
| 430 |
+
"step": 27500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.752206381534283,
|
| 434 |
+
"grad_norm": 0.04381977394223213,
|
| 435 |
+
"learning_rate": 5.582839809331215e-06,
|
| 436 |
+
"loss": 0.1061,
|
| 437 |
+
"step": 28000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.8370672097759675,
|
| 441 |
+
"grad_norm": 8.241320610046387,
|
| 442 |
+
"learning_rate": 5.492561028455872e-06,
|
| 443 |
+
"loss": 0.1008,
|
| 444 |
+
"step": 28500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.921928038017651,
|
| 448 |
+
"grad_norm": 34.80970001220703,
|
| 449 |
+
"learning_rate": 5.402282247580529e-06,
|
| 450 |
+
"loss": 0.1069,
|
| 451 |
+
"step": 29000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 5.0,
|
| 455 |
+
"eval_accuracy": 0.8975656324582338,
|
| 456 |
+
"eval_loss": 0.5583851933479309,
|
| 457 |
+
"eval_runtime": 7.324,
|
| 458 |
+
"eval_samples_per_second": 1430.238,
|
| 459 |
+
"eval_steps_per_second": 89.433,
|
| 460 |
+
"step": 29460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.006788866259335,
|
| 464 |
+
"grad_norm": 0.9499875903129578,
|
| 465 |
+
"learning_rate": 5.312003466705185e-06,
|
| 466 |
+
"loss": 0.0916,
|
| 467 |
+
"step": 29500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.091649694501018,
|
| 471 |
+
"grad_norm": 4.829331398010254,
|
| 472 |
+
"learning_rate": 5.221724685829843e-06,
|
| 473 |
+
"loss": 0.0595,
|
| 474 |
+
"step": 30000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.176510522742702,
|
| 478 |
+
"grad_norm": 0.05401415750384331,
|
| 479 |
+
"learning_rate": 5.1314459049545e-06,
|
| 480 |
+
"loss": 0.0546,
|
| 481 |
+
"step": 30500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.261371350984386,
|
| 485 |
+
"grad_norm": 69.64205169677734,
|
| 486 |
+
"learning_rate": 5.041167124079157e-06,
|
| 487 |
+
"loss": 0.0682,
|
| 488 |
+
"step": 31000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.346232179226069,
|
| 492 |
+
"grad_norm": 0.0484016053378582,
|
| 493 |
+
"learning_rate": 4.950888343203813e-06,
|
| 494 |
+
"loss": 0.0764,
|
| 495 |
+
"step": 31500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.431093007467753,
|
| 499 |
+
"grad_norm": 0.030390514060854912,
|
| 500 |
+
"learning_rate": 4.86060956232847e-06,
|
| 501 |
+
"loss": 0.0721,
|
| 502 |
+
"step": 32000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.515953835709436,
|
| 506 |
+
"grad_norm": 0.0501534603536129,
|
| 507 |
+
"learning_rate": 4.7703307814531275e-06,
|
| 508 |
+
"loss": 0.0649,
|
| 509 |
+
"step": 32500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.6008146639511205,
|
| 513 |
+
"grad_norm": 0.20272402465343475,
|
| 514 |
+
"learning_rate": 4.680052000577785e-06,
|
| 515 |
+
"loss": 0.0672,
|
| 516 |
+
"step": 33000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 5.685675492192804,
|
| 520 |
+
"grad_norm": 0.06525377184152603,
|
| 521 |
+
"learning_rate": 4.589773219702441e-06,
|
| 522 |
+
"loss": 0.0712,
|
| 523 |
+
"step": 33500
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 5.770536320434488,
|
| 527 |
+
"grad_norm": 0.10100004076957703,
|
| 528 |
+
"learning_rate": 4.499494438827098e-06,
|
| 529 |
+
"loss": 0.0682,
|
| 530 |
+
"step": 34000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.855397148676171,
|
| 534 |
+
"grad_norm": 0.05232209339737892,
|
| 535 |
+
"learning_rate": 4.4092156579517554e-06,
|
| 536 |
+
"loss": 0.0703,
|
| 537 |
+
"step": 34500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.940257976917855,
|
| 541 |
+
"grad_norm": 0.08663167804479599,
|
| 542 |
+
"learning_rate": 4.318936877076413e-06,
|
| 543 |
+
"loss": 0.0632,
|
| 544 |
+
"step": 35000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 6.0,
|
| 548 |
+
"eval_accuracy": 0.8988066825775657,
|
| 549 |
+
"eval_loss": 0.6263558864593506,
|
| 550 |
+
"eval_runtime": 7.3364,
|
| 551 |
+
"eval_samples_per_second": 1427.812,
|
| 552 |
+
"eval_steps_per_second": 89.281,
|
| 553 |
+
"step": 35352
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.025118805159538,
|
| 557 |
+
"grad_norm": 0.05715448781847954,
|
| 558 |
+
"learning_rate": 4.228658096201069e-06,
|
| 559 |
+
"loss": 0.0631,
|
| 560 |
+
"step": 35500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.109979633401222,
|
| 564 |
+
"grad_norm": 0.039631437510252,
|
| 565 |
+
"learning_rate": 4.138379315325726e-06,
|
| 566 |
+
"loss": 0.0498,
|
| 567 |
+
"step": 36000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.194840461642905,
|
| 571 |
+
"grad_norm": 2.1679649353027344,
|
| 572 |
+
"learning_rate": 4.048100534450383e-06,
|
| 573 |
+
"loss": 0.0462,
|
| 574 |
+
"step": 36500
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 6.279701289884589,
|
| 578 |
+
"grad_norm": 0.71781325340271,
|
| 579 |
+
"learning_rate": 3.9578217535750406e-06,
|
| 580 |
+
"loss": 0.0517,
|
| 581 |
+
"step": 37000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 6.364562118126273,
|
| 585 |
+
"grad_norm": 0.03331080824136734,
|
| 586 |
+
"learning_rate": 3.867542972699697e-06,
|
| 587 |
+
"loss": 0.0645,
|
| 588 |
+
"step": 37500
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 6.449422946367957,
|
| 592 |
+
"grad_norm": 0.03804829716682434,
|
| 593 |
+
"learning_rate": 3.777264191824354e-06,
|
| 594 |
+
"loss": 0.0467,
|
| 595 |
+
"step": 38000
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.53428377460964,
|
| 599 |
+
"grad_norm": 0.013244557194411755,
|
| 600 |
+
"learning_rate": 3.686985410949011e-06,
|
| 601 |
+
"loss": 0.0459,
|
| 602 |
+
"step": 38500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.619144602851324,
|
| 606 |
+
"grad_norm": 0.3420211374759674,
|
| 607 |
+
"learning_rate": 3.596706630073668e-06,
|
| 608 |
+
"loss": 0.0444,
|
| 609 |
+
"step": 39000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.704005431093007,
|
| 613 |
+
"grad_norm": 0.08247170597314835,
|
| 614 |
+
"learning_rate": 3.5064278491983244e-06,
|
| 615 |
+
"loss": 0.0551,
|
| 616 |
+
"step": 39500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.788866259334691,
|
| 620 |
+
"grad_norm": 0.3416644334793091,
|
| 621 |
+
"learning_rate": 3.4161490683229816e-06,
|
| 622 |
+
"loss": 0.0431,
|
| 623 |
+
"step": 40000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 6.873727087576375,
|
| 627 |
+
"grad_norm": 0.025737063959240913,
|
| 628 |
+
"learning_rate": 3.3258702874476384e-06,
|
| 629 |
+
"loss": 0.0446,
|
| 630 |
+
"step": 40500
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.958587915818058,
|
| 634 |
+
"grad_norm": 0.020823437720537186,
|
| 635 |
+
"learning_rate": 3.2355915065722956e-06,
|
| 636 |
+
"loss": 0.0524,
|
| 637 |
+
"step": 41000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 7.0,
|
| 641 |
+
"eval_accuracy": 0.8997613365155132,
|
| 642 |
+
"eval_loss": 0.6955571174621582,
|
| 643 |
+
"eval_runtime": 7.3233,
|
| 644 |
+
"eval_samples_per_second": 1430.365,
|
| 645 |
+
"eval_steps_per_second": 89.44,
|
| 646 |
+
"step": 41244
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.043448744059742,
|
| 650 |
+
"grad_norm": 0.006505677942186594,
|
| 651 |
+
"learning_rate": 3.1453127256969523e-06,
|
| 652 |
+
"loss": 0.0398,
|
| 653 |
+
"step": 41500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.128309572301426,
|
| 657 |
+
"grad_norm": 0.1108703762292862,
|
| 658 |
+
"learning_rate": 3.055033944821609e-06,
|
| 659 |
+
"loss": 0.0293,
|
| 660 |
+
"step": 42000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.21317040054311,
|
| 664 |
+
"grad_norm": 0.038421932607889175,
|
| 665 |
+
"learning_rate": 2.9647551639462663e-06,
|
| 666 |
+
"loss": 0.0281,
|
| 667 |
+
"step": 42500
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.298031228784793,
|
| 671 |
+
"grad_norm": 0.005532279144972563,
|
| 672 |
+
"learning_rate": 2.874476383070923e-06,
|
| 673 |
+
"loss": 0.0351,
|
| 674 |
+
"step": 43000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.382892057026477,
|
| 678 |
+
"grad_norm": 0.07227976620197296,
|
| 679 |
+
"learning_rate": 2.7841976021955803e-06,
|
| 680 |
+
"loss": 0.0349,
|
| 681 |
+
"step": 43500
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.46775288526816,
|
| 685 |
+
"grad_norm": 0.013657882809638977,
|
| 686 |
+
"learning_rate": 2.693918821320237e-06,
|
| 687 |
+
"loss": 0.0344,
|
| 688 |
+
"step": 44000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.552613713509844,
|
| 692 |
+
"grad_norm": 0.008297976106405258,
|
| 693 |
+
"learning_rate": 2.6036400404448942e-06,
|
| 694 |
+
"loss": 0.0243,
|
| 695 |
+
"step": 44500
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.637474541751527,
|
| 699 |
+
"grad_norm": 0.0019395159324631095,
|
| 700 |
+
"learning_rate": 2.513361259569551e-06,
|
| 701 |
+
"loss": 0.0306,
|
| 702 |
+
"step": 45000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 7.722335369993211,
|
| 706 |
+
"grad_norm": 27.918182373046875,
|
| 707 |
+
"learning_rate": 2.4230824786942078e-06,
|
| 708 |
+
"loss": 0.0348,
|
| 709 |
+
"step": 45500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 7.8071961982348945,
|
| 713 |
+
"grad_norm": 47.90977478027344,
|
| 714 |
+
"learning_rate": 2.332803697818865e-06,
|
| 715 |
+
"loss": 0.0383,
|
| 716 |
+
"step": 46000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 7.892057026476579,
|
| 720 |
+
"grad_norm": 0.008037786930799484,
|
| 721 |
+
"learning_rate": 2.2425249169435217e-06,
|
| 722 |
+
"loss": 0.0276,
|
| 723 |
+
"step": 46500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.976917854718262,
|
| 727 |
+
"grad_norm": 0.055451564490795135,
|
| 728 |
+
"learning_rate": 2.152246136068179e-06,
|
| 729 |
+
"loss": 0.0342,
|
| 730 |
+
"step": 47000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 8.0,
|
| 734 |
+
"eval_accuracy": 0.8987112171837709,
|
| 735 |
+
"eval_loss": 0.7324450612068176,
|
| 736 |
+
"eval_runtime": 7.3375,
|
| 737 |
+
"eval_samples_per_second": 1427.589,
|
| 738 |
+
"eval_steps_per_second": 89.267,
|
| 739 |
+
"step": 47136
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 8.061778682959945,
|
| 743 |
+
"grad_norm": 0.020909298211336136,
|
| 744 |
+
"learning_rate": 2.0619673551928357e-06,
|
| 745 |
+
"loss": 0.0251,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 8.146639511201629,
|
| 750 |
+
"grad_norm": 0.00416733231395483,
|
| 751 |
+
"learning_rate": 1.9716885743174925e-06,
|
| 752 |
+
"loss": 0.0245,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.231500339443313,
|
| 757 |
+
"grad_norm": 0.004132431466132402,
|
| 758 |
+
"learning_rate": 1.8814097934421497e-06,
|
| 759 |
+
"loss": 0.0188,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 8.316361167684997,
|
| 764 |
+
"grad_norm": 0.003154342994093895,
|
| 765 |
+
"learning_rate": 1.7911310125668062e-06,
|
| 766 |
+
"loss": 0.0224,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.40122199592668,
|
| 771 |
+
"grad_norm": 9.123353958129883,
|
| 772 |
+
"learning_rate": 1.7008522316914632e-06,
|
| 773 |
+
"loss": 0.0259,
|
| 774 |
+
"step": 49500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 8.486082824168363,
|
| 778 |
+
"grad_norm": 0.9837649464607239,
|
| 779 |
+
"learning_rate": 1.6105734508161202e-06,
|
| 780 |
+
"loss": 0.0234,
|
| 781 |
+
"step": 50000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 8.570943652410048,
|
| 785 |
+
"grad_norm": 0.018608735874295235,
|
| 786 |
+
"learning_rate": 1.5202946699407772e-06,
|
| 787 |
+
"loss": 0.0249,
|
| 788 |
+
"step": 50500
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 8.655804480651732,
|
| 792 |
+
"grad_norm": 0.1524048149585724,
|
| 793 |
+
"learning_rate": 1.4300158890654341e-06,
|
| 794 |
+
"loss": 0.0277,
|
| 795 |
+
"step": 51000
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.740665308893416,
|
| 799 |
+
"grad_norm": 0.0041467552073299885,
|
| 800 |
+
"learning_rate": 1.3397371081900911e-06,
|
| 801 |
+
"loss": 0.0293,
|
| 802 |
+
"step": 51500
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.825526137135098,
|
| 806 |
+
"grad_norm": 0.030655721202492714,
|
| 807 |
+
"learning_rate": 1.249458327314748e-06,
|
| 808 |
+
"loss": 0.0239,
|
| 809 |
+
"step": 52000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.910386965376782,
|
| 813 |
+
"grad_norm": 0.008702299557626247,
|
| 814 |
+
"learning_rate": 1.1591795464394049e-06,
|
| 815 |
+
"loss": 0.0223,
|
| 816 |
+
"step": 52500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 8.995247793618466,
|
| 820 |
+
"grad_norm": 0.09563937038183212,
|
| 821 |
+
"learning_rate": 1.0689007655640619e-06,
|
| 822 |
+
"loss": 0.0259,
|
| 823 |
+
"step": 53000
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 9.0,
|
| 827 |
+
"eval_accuracy": 0.8990930787589498,
|
| 828 |
+
"eval_loss": 0.7912434935569763,
|
| 829 |
+
"eval_runtime": 7.3146,
|
| 830 |
+
"eval_samples_per_second": 1432.066,
|
| 831 |
+
"eval_steps_per_second": 89.547,
|
| 832 |
+
"step": 53028
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 9.08010862186015,
|
| 836 |
+
"grad_norm": 0.0016417688457295299,
|
| 837 |
+
"learning_rate": 9.786219846887188e-07,
|
| 838 |
+
"loss": 0.0158,
|
| 839 |
+
"step": 53500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.164969450101832,
|
| 843 |
+
"grad_norm": 0.007821256294846535,
|
| 844 |
+
"learning_rate": 8.883432038133757e-07,
|
| 845 |
+
"loss": 0.0156,
|
| 846 |
+
"step": 54000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 9.249830278343516,
|
| 850 |
+
"grad_norm": 0.0008865104173310101,
|
| 851 |
+
"learning_rate": 7.980644229380327e-07,
|
| 852 |
+
"loss": 0.0124,
|
| 853 |
+
"step": 54500
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 9.3346911065852,
|
| 857 |
+
"grad_norm": 70.18607330322266,
|
| 858 |
+
"learning_rate": 7.077856420626897e-07,
|
| 859 |
+
"loss": 0.0204,
|
| 860 |
+
"step": 55000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 9.419551934826885,
|
| 864 |
+
"grad_norm": 0.005583506077528,
|
| 865 |
+
"learning_rate": 6.175068611873466e-07,
|
| 866 |
+
"loss": 0.0185,
|
| 867 |
+
"step": 55500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 9.504412763068567,
|
| 871 |
+
"grad_norm": 0.003502935403957963,
|
| 872 |
+
"learning_rate": 5.272280803120034e-07,
|
| 873 |
+
"loss": 0.0207,
|
| 874 |
+
"step": 56000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 9.589273591310251,
|
| 878 |
+
"grad_norm": 0.005448056850582361,
|
| 879 |
+
"learning_rate": 4.3694929943666047e-07,
|
| 880 |
+
"loss": 0.0247,
|
| 881 |
+
"step": 56500
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 9.674134419551935,
|
| 885 |
+
"grad_norm": 0.049458473920822144,
|
| 886 |
+
"learning_rate": 3.4667051856131735e-07,
|
| 887 |
+
"loss": 0.0175,
|
| 888 |
+
"step": 57000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 9.758995247793619,
|
| 892 |
+
"grad_norm": 19.99628448486328,
|
| 893 |
+
"learning_rate": 2.5639173768597433e-07,
|
| 894 |
+
"loss": 0.0205,
|
| 895 |
+
"step": 57500
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 9.843856076035301,
|
| 899 |
+
"grad_norm": 0.025633709505200386,
|
| 900 |
+
"learning_rate": 1.6611295681063126e-07,
|
| 901 |
+
"loss": 0.0197,
|
| 902 |
+
"step": 58000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.928716904276985,
|
| 906 |
+
"grad_norm": 1.0917688608169556,
|
| 907 |
+
"learning_rate": 7.583417593528817e-08,
|
| 908 |
+
"loss": 0.0187,
|
| 909 |
+
"step": 58500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 10.0,
|
| 913 |
+
"eval_accuracy": 0.8998568019093078,
|
| 914 |
+
"eval_loss": 0.8246611952781677,
|
| 915 |
+
"eval_runtime": 7.324,
|
| 916 |
+
"eval_samples_per_second": 1430.235,
|
| 917 |
+
"eval_steps_per_second": 89.432,
|
| 918 |
+
"step": 58920
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 10.0,
|
| 922 |
+
"step": 58920,
|
| 923 |
+
"total_flos": 4.238588368543968e+16,
|
| 924 |
+
"train_loss": 0.1337772551961926,
|
| 925 |
+
"train_runtime": 3247.3369,
|
| 926 |
+
"train_samples_per_second": 290.293,
|
| 927 |
+
"train_steps_per_second": 18.144
|
| 928 |
+
}
|
| 929 |
+
],
|
| 930 |
+
"logging_steps": 500,
|
| 931 |
+
"max_steps": 58920,
|
| 932 |
+
"num_input_tokens_seen": 0,
|
| 933 |
+
"num_train_epochs": 10,
|
| 934 |
+
"save_steps": 500,
|
| 935 |
+
"stateful_callbacks": {
|
| 936 |
+
"TrainerControl": {
|
| 937 |
+
"args": {
|
| 938 |
+
"should_epoch_stop": false,
|
| 939 |
+
"should_evaluate": false,
|
| 940 |
+
"should_log": false,
|
| 941 |
+
"should_save": true,
|
| 942 |
+
"should_training_stop": true
|
| 943 |
+
},
|
| 944 |
+
"attributes": {}
|
| 945 |
+
}
|
| 946 |
+
},
|
| 947 |
+
"total_flos": 4.238588368543968e+16,
|
| 948 |
+
"train_batch_size": 16,
|
| 949 |
+
"trial_name": null,
|
| 950 |
+
"trial_params": null
|
| 951 |
+
}
|
qnli/bert-base-uncased_lr1e-05/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bf4a88696b05928e196da98c13cb34e3874e8bda302abd4177ce01778d3df39
|
| 3 |
+
size 5240
|
qnli/bert-base-uncased_lr1e-05/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-base_lr1e-05/classifier_head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f78b473d61289116d47f32f8323270ea6eaf2b0587e3413ec348da8ff300b8fc
|
| 3 |
+
size 2371920
|
qnli/roberta-base_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/shikexuan/nlu_model/roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.45.2",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
qnli/roberta-base_lr1e-05/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-base_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb444f9065c6464b582e9b25cba871566b41dc14253d6679d1c0c0101c64f04b
|
| 3 |
+
size 498612824
|
qnli/roberta-base_lr1e-05/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
qnli/roberta-base_lr1e-05/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-base_lr1e-05/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"mask_token": "<mask>",
|
| 51 |
+
"model_max_length": 512,
|
| 52 |
+
"pad_token": "<pad>",
|
| 53 |
+
"sep_token": "</s>",
|
| 54 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 55 |
+
"trim_offsets": true,
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
qnli/roberta-base_lr1e-05/trainer_state.json
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9155131264916467,
|
| 3 |
+
"best_model_checkpoint": "./nlu_finetuned_models/qnli/roberta-base_lr1e-05/checkpoint-58920",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 58920,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08486082824168364,
|
| 13 |
+
"grad_norm": 1.849210500717163,
|
| 14 |
+
"learning_rate": 1.4140271493212672e-06,
|
| 15 |
+
"loss": 0.7001,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1697216564833673,
|
| 20 |
+
"grad_norm": 16.44700050354004,
|
| 21 |
+
"learning_rate": 2.8280542986425343e-06,
|
| 22 |
+
"loss": 0.5552,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.2545824847250509,
|
| 27 |
+
"grad_norm": 23.17144012451172,
|
| 28 |
+
"learning_rate": 4.242081447963801e-06,
|
| 29 |
+
"loss": 0.4292,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3394433129667346,
|
| 34 |
+
"grad_norm": 27.956209182739258,
|
| 35 |
+
"learning_rate": 5.656108597285069e-06,
|
| 36 |
+
"loss": 0.3972,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4243041412084182,
|
| 41 |
+
"grad_norm": 26.34242057800293,
|
| 42 |
+
"learning_rate": 7.070135746606336e-06,
|
| 43 |
+
"loss": 0.3795,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5091649694501018,
|
| 48 |
+
"grad_norm": 27.0740966796875,
|
| 49 |
+
"learning_rate": 8.484162895927603e-06,
|
| 50 |
+
"loss": 0.3601,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.5940257976917854,
|
| 55 |
+
"grad_norm": 18.370553970336914,
|
| 56 |
+
"learning_rate": 9.89819004524887e-06,
|
| 57 |
+
"loss": 0.3498,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6788866259334692,
|
| 62 |
+
"grad_norm": 26.062349319458008,
|
| 63 |
+
"learning_rate": 9.916221291347683e-06,
|
| 64 |
+
"loss": 0.3305,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.7637474541751528,
|
| 69 |
+
"grad_norm": 20.309837341308594,
|
| 70 |
+
"learning_rate": 9.825942510472339e-06,
|
| 71 |
+
"loss": 0.3233,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.8486082824168364,
|
| 76 |
+
"grad_norm": 17.478370666503906,
|
| 77 |
+
"learning_rate": 9.735663729596995e-06,
|
| 78 |
+
"loss": 0.299,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.93346911065852,
|
| 83 |
+
"grad_norm": 31.12592315673828,
|
| 84 |
+
"learning_rate": 9.645384948721653e-06,
|
| 85 |
+
"loss": 0.3009,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.0,
|
| 90 |
+
"eval_accuracy": 0.9017661097852029,
|
| 91 |
+
"eval_loss": 0.25566375255584717,
|
| 92 |
+
"eval_runtime": 7.6504,
|
| 93 |
+
"eval_samples_per_second": 1369.202,
|
| 94 |
+
"eval_steps_per_second": 85.616,
|
| 95 |
+
"step": 5892
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.0183299389002036,
|
| 99 |
+
"grad_norm": 14.976820945739746,
|
| 100 |
+
"learning_rate": 9.55510616784631e-06,
|
| 101 |
+
"loss": 0.2757,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1031907671418872,
|
| 106 |
+
"grad_norm": 14.064760208129883,
|
| 107 |
+
"learning_rate": 9.464827386970968e-06,
|
| 108 |
+
"loss": 0.274,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1880515953835709,
|
| 113 |
+
"grad_norm": 13.457983016967773,
|
| 114 |
+
"learning_rate": 9.374548606095624e-06,
|
| 115 |
+
"loss": 0.2582,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.2729124236252547,
|
| 120 |
+
"grad_norm": 22.956470489501953,
|
| 121 |
+
"learning_rate": 9.28426982522028e-06,
|
| 122 |
+
"loss": 0.2579,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.3577732518669383,
|
| 127 |
+
"grad_norm": 15.31163215637207,
|
| 128 |
+
"learning_rate": 9.193991044344937e-06,
|
| 129 |
+
"loss": 0.2531,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.442634080108622,
|
| 134 |
+
"grad_norm": 24.784841537475586,
|
| 135 |
+
"learning_rate": 9.103712263469595e-06,
|
| 136 |
+
"loss": 0.2701,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.5274949083503055,
|
| 141 |
+
"grad_norm": 15.537549018859863,
|
| 142 |
+
"learning_rate": 9.013433482594251e-06,
|
| 143 |
+
"loss": 0.2472,
|
| 144 |
+
"step": 9000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.6123557365919892,
|
| 148 |
+
"grad_norm": 10.927321434020996,
|
| 149 |
+
"learning_rate": 8.92315470171891e-06,
|
| 150 |
+
"loss": 0.2515,
|
| 151 |
+
"step": 9500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.6972165648336728,
|
| 155 |
+
"grad_norm": 7.331003665924072,
|
| 156 |
+
"learning_rate": 8.832875920843566e-06,
|
| 157 |
+
"loss": 0.2418,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.7820773930753564,
|
| 162 |
+
"grad_norm": 18.43935203552246,
|
| 163 |
+
"learning_rate": 8.742597139968224e-06,
|
| 164 |
+
"loss": 0.2546,
|
| 165 |
+
"step": 10500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.86693822131704,
|
| 169 |
+
"grad_norm": 25.023645401000977,
|
| 170 |
+
"learning_rate": 8.65231835909288e-06,
|
| 171 |
+
"loss": 0.2477,
|
| 172 |
+
"step": 11000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.9517990495587236,
|
| 176 |
+
"grad_norm": 7.594945907592773,
|
| 177 |
+
"learning_rate": 8.562039578217536e-06,
|
| 178 |
+
"loss": 0.2505,
|
| 179 |
+
"step": 11500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_accuracy": 0.9132219570405727,
|
| 184 |
+
"eval_loss": 0.2617391049861908,
|
| 185 |
+
"eval_runtime": 7.6353,
|
| 186 |
+
"eval_samples_per_second": 1371.912,
|
| 187 |
+
"eval_steps_per_second": 85.785,
|
| 188 |
+
"step": 11784
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.0366598778004072,
|
| 192 |
+
"grad_norm": 8.508316040039062,
|
| 193 |
+
"learning_rate": 8.471760797342193e-06,
|
| 194 |
+
"loss": 0.2221,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.121520706042091,
|
| 199 |
+
"grad_norm": 15.623618125915527,
|
| 200 |
+
"learning_rate": 8.381482016466849e-06,
|
| 201 |
+
"loss": 0.197,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.2063815342837745,
|
| 206 |
+
"grad_norm": 0.13940224051475525,
|
| 207 |
+
"learning_rate": 8.291203235591507e-06,
|
| 208 |
+
"loss": 0.2096,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.291242362525458,
|
| 213 |
+
"grad_norm": 14.592768669128418,
|
| 214 |
+
"learning_rate": 8.200924454716163e-06,
|
| 215 |
+
"loss": 0.2117,
|
| 216 |
+
"step": 13500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.3761031907671417,
|
| 220 |
+
"grad_norm": 25.236705780029297,
|
| 221 |
+
"learning_rate": 8.110645673840822e-06,
|
| 222 |
+
"loss": 0.2055,
|
| 223 |
+
"step": 14000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.4609640190088253,
|
| 227 |
+
"grad_norm": 10.421687126159668,
|
| 228 |
+
"learning_rate": 8.020366892965478e-06,
|
| 229 |
+
"loss": 0.2158,
|
| 230 |
+
"step": 14500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.5458248472505094,
|
| 234 |
+
"grad_norm": 18.31736183166504,
|
| 235 |
+
"learning_rate": 7.930088112090136e-06,
|
| 236 |
+
"loss": 0.2171,
|
| 237 |
+
"step": 15000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.6306856754921926,
|
| 241 |
+
"grad_norm": 51.31163024902344,
|
| 242 |
+
"learning_rate": 7.839809331214792e-06,
|
| 243 |
+
"loss": 0.2043,
|
| 244 |
+
"step": 15500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.7155465037338766,
|
| 248 |
+
"grad_norm": 44.53534698486328,
|
| 249 |
+
"learning_rate": 7.749530550339449e-06,
|
| 250 |
+
"loss": 0.2017,
|
| 251 |
+
"step": 16000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.8004073319755602,
|
| 255 |
+
"grad_norm": 6.736083030700684,
|
| 256 |
+
"learning_rate": 7.659251769464105e-06,
|
| 257 |
+
"loss": 0.203,
|
| 258 |
+
"step": 16500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.885268160217244,
|
| 262 |
+
"grad_norm": 21.882688522338867,
|
| 263 |
+
"learning_rate": 7.568972988588763e-06,
|
| 264 |
+
"loss": 0.2073,
|
| 265 |
+
"step": 17000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.9701289884589275,
|
| 269 |
+
"grad_norm": 45.52244567871094,
|
| 270 |
+
"learning_rate": 7.478694207713419e-06,
|
| 271 |
+
"loss": 0.2094,
|
| 272 |
+
"step": 17500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.0,
|
| 276 |
+
"eval_accuracy": 0.9143675417661098,
|
| 277 |
+
"eval_loss": 0.2764427363872528,
|
| 278 |
+
"eval_runtime": 7.5969,
|
| 279 |
+
"eval_samples_per_second": 1378.853,
|
| 280 |
+
"eval_steps_per_second": 86.219,
|
| 281 |
+
"step": 17676
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.054989816700611,
|
| 285 |
+
"grad_norm": 10.752288818359375,
|
| 286 |
+
"learning_rate": 7.388415426838077e-06,
|
| 287 |
+
"loss": 0.1919,
|
| 288 |
+
"step": 18000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.1398506449422947,
|
| 292 |
+
"grad_norm": 41.408321380615234,
|
| 293 |
+
"learning_rate": 7.298136645962734e-06,
|
| 294 |
+
"loss": 0.1649,
|
| 295 |
+
"step": 18500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.2247114731839783,
|
| 299 |
+
"grad_norm": 66.92359924316406,
|
| 300 |
+
"learning_rate": 7.207857865087391e-06,
|
| 301 |
+
"loss": 0.185,
|
| 302 |
+
"step": 19000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.309572301425662,
|
| 306 |
+
"grad_norm": 4.060750484466553,
|
| 307 |
+
"learning_rate": 7.117579084212047e-06,
|
| 308 |
+
"loss": 0.1792,
|
| 309 |
+
"step": 19500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.3944331296673456,
|
| 313 |
+
"grad_norm": 0.46891507506370544,
|
| 314 |
+
"learning_rate": 7.027300303336704e-06,
|
| 315 |
+
"loss": 0.1735,
|
| 316 |
+
"step": 20000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.479293957909029,
|
| 320 |
+
"grad_norm": 38.78276443481445,
|
| 321 |
+
"learning_rate": 6.937021522461362e-06,
|
| 322 |
+
"loss": 0.1778,
|
| 323 |
+
"step": 20500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.564154786150713,
|
| 327 |
+
"grad_norm": 1.1472349166870117,
|
| 328 |
+
"learning_rate": 6.846742741586018e-06,
|
| 329 |
+
"loss": 0.1815,
|
| 330 |
+
"step": 21000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.6490156143923964,
|
| 334 |
+
"grad_norm": 25.915781021118164,
|
| 335 |
+
"learning_rate": 6.756463960710675e-06,
|
| 336 |
+
"loss": 0.1775,
|
| 337 |
+
"step": 21500
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.73387644263408,
|
| 341 |
+
"grad_norm": 1.1724231243133545,
|
| 342 |
+
"learning_rate": 6.6661851798353315e-06,
|
| 343 |
+
"loss": 0.1922,
|
| 344 |
+
"step": 22000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.8187372708757636,
|
| 348 |
+
"grad_norm": 0.4711698591709137,
|
| 349 |
+
"learning_rate": 6.57590639895999e-06,
|
| 350 |
+
"loss": 0.1816,
|
| 351 |
+
"step": 22500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.9035980991174473,
|
| 355 |
+
"grad_norm": 25.3336238861084,
|
| 356 |
+
"learning_rate": 6.485627618084646e-06,
|
| 357 |
+
"loss": 0.1871,
|
| 358 |
+
"step": 23000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.988458927359131,
|
| 362 |
+
"grad_norm": 6.833723068237305,
|
| 363 |
+
"learning_rate": 6.395348837209303e-06,
|
| 364 |
+
"loss": 0.1796,
|
| 365 |
+
"step": 23500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.0,
|
| 369 |
+
"eval_accuracy": 0.9118854415274463,
|
| 370 |
+
"eval_loss": 0.41552412509918213,
|
| 371 |
+
"eval_runtime": 7.6739,
|
| 372 |
+
"eval_samples_per_second": 1365.009,
|
| 373 |
+
"eval_steps_per_second": 85.354,
|
| 374 |
+
"step": 23568
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.0733197556008145,
|
| 378 |
+
"grad_norm": 0.6186419725418091,
|
| 379 |
+
"learning_rate": 6.3050700563339595e-06,
|
| 380 |
+
"loss": 0.1287,
|
| 381 |
+
"step": 24000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.158180583842499,
|
| 385 |
+
"grad_norm": 0.3264002799987793,
|
| 386 |
+
"learning_rate": 6.214791275458617e-06,
|
| 387 |
+
"loss": 0.141,
|
| 388 |
+
"step": 24500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.243041412084182,
|
| 392 |
+
"grad_norm": 29.884204864501953,
|
| 393 |
+
"learning_rate": 6.124512494583274e-06,
|
| 394 |
+
"loss": 0.1419,
|
| 395 |
+
"step": 25000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.327902240325866,
|
| 399 |
+
"grad_norm": 83.65303039550781,
|
| 400 |
+
"learning_rate": 6.034233713707931e-06,
|
| 401 |
+
"loss": 0.1401,
|
| 402 |
+
"step": 25500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.412763068567549,
|
| 406 |
+
"grad_norm": 59.16463088989258,
|
| 407 |
+
"learning_rate": 5.943954932832587e-06,
|
| 408 |
+
"loss": 0.1575,
|
| 409 |
+
"step": 26000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.497623896809233,
|
| 413 |
+
"grad_norm": 76.32769775390625,
|
| 414 |
+
"learning_rate": 5.853676151957245e-06,
|
| 415 |
+
"loss": 0.1374,
|
| 416 |
+
"step": 26500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.582484725050916,
|
| 420 |
+
"grad_norm": 40.29171371459961,
|
| 421 |
+
"learning_rate": 5.763397371081901e-06,
|
| 422 |
+
"loss": 0.1437,
|
| 423 |
+
"step": 27000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.6673455532926,
|
| 427 |
+
"grad_norm": 0.6104162335395813,
|
| 428 |
+
"learning_rate": 5.673118590206559e-06,
|
| 429 |
+
"loss": 0.1657,
|
| 430 |
+
"step": 27500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.752206381534283,
|
| 434 |
+
"grad_norm": 8.045665740966797,
|
| 435 |
+
"learning_rate": 5.582839809331215e-06,
|
| 436 |
+
"loss": 0.1476,
|
| 437 |
+
"step": 28000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.8370672097759675,
|
| 441 |
+
"grad_norm": 2.300698757171631,
|
| 442 |
+
"learning_rate": 5.492561028455872e-06,
|
| 443 |
+
"loss": 0.146,
|
| 444 |
+
"step": 28500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.921928038017651,
|
| 448 |
+
"grad_norm": 92.35627746582031,
|
| 449 |
+
"learning_rate": 5.402282247580529e-06,
|
| 450 |
+
"loss": 0.153,
|
| 451 |
+
"step": 29000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 5.0,
|
| 455 |
+
"eval_accuracy": 0.9125536992840095,
|
| 456 |
+
"eval_loss": 0.41520774364471436,
|
| 457 |
+
"eval_runtime": 7.6176,
|
| 458 |
+
"eval_samples_per_second": 1375.098,
|
| 459 |
+
"eval_steps_per_second": 85.985,
|
| 460 |
+
"step": 29460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.006788866259335,
|
| 464 |
+
"grad_norm": 3.93399977684021,
|
| 465 |
+
"learning_rate": 5.312003466705185e-06,
|
| 466 |
+
"loss": 0.1495,
|
| 467 |
+
"step": 29500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.091649694501018,
|
| 471 |
+
"grad_norm": 0.2336718887090683,
|
| 472 |
+
"learning_rate": 5.221724685829843e-06,
|
| 473 |
+
"loss": 0.1054,
|
| 474 |
+
"step": 30000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.176510522742702,
|
| 478 |
+
"grad_norm": 0.31675928831100464,
|
| 479 |
+
"learning_rate": 5.1314459049545e-06,
|
| 480 |
+
"loss": 0.1151,
|
| 481 |
+
"step": 30500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.261371350984386,
|
| 485 |
+
"grad_norm": 0.19725990295410156,
|
| 486 |
+
"learning_rate": 5.041167124079157e-06,
|
| 487 |
+
"loss": 0.1163,
|
| 488 |
+
"step": 31000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.346232179226069,
|
| 492 |
+
"grad_norm": 0.11642979085445404,
|
| 493 |
+
"learning_rate": 4.950888343203813e-06,
|
| 494 |
+
"loss": 0.1084,
|
| 495 |
+
"step": 31500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.431093007467753,
|
| 499 |
+
"grad_norm": 31.47455406188965,
|
| 500 |
+
"learning_rate": 4.86060956232847e-06,
|
| 501 |
+
"loss": 0.1081,
|
| 502 |
+
"step": 32000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.515953835709436,
|
| 506 |
+
"grad_norm": 427.69683837890625,
|
| 507 |
+
"learning_rate": 4.7703307814531275e-06,
|
| 508 |
+
"loss": 0.1114,
|
| 509 |
+
"step": 32500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.6008146639511205,
|
| 513 |
+
"grad_norm": 22.574277877807617,
|
| 514 |
+
"learning_rate": 4.680052000577785e-06,
|
| 515 |
+
"loss": 0.1181,
|
| 516 |
+
"step": 33000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 5.685675492192804,
|
| 520 |
+
"grad_norm": 0.2410123497247696,
|
| 521 |
+
"learning_rate": 4.589773219702441e-06,
|
| 522 |
+
"loss": 0.1168,
|
| 523 |
+
"step": 33500
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 5.770536320434488,
|
| 527 |
+
"grad_norm": 0.6157441735267639,
|
| 528 |
+
"learning_rate": 4.499494438827098e-06,
|
| 529 |
+
"loss": 0.1112,
|
| 530 |
+
"step": 34000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.855397148676171,
|
| 534 |
+
"grad_norm": 0.3697001039981842,
|
| 535 |
+
"learning_rate": 4.4092156579517554e-06,
|
| 536 |
+
"loss": 0.1261,
|
| 537 |
+
"step": 34500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.940257976917855,
|
| 541 |
+
"grad_norm": 0.13417810201644897,
|
| 542 |
+
"learning_rate": 4.318936877076413e-06,
|
| 543 |
+
"loss": 0.1203,
|
| 544 |
+
"step": 35000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 6.0,
|
| 548 |
+
"eval_accuracy": 0.9151312649164678,
|
| 549 |
+
"eval_loss": 0.4638233184814453,
|
| 550 |
+
"eval_runtime": 7.6011,
|
| 551 |
+
"eval_samples_per_second": 1378.094,
|
| 552 |
+
"eval_steps_per_second": 86.172,
|
| 553 |
+
"step": 35352
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.025118805159538,
|
| 557 |
+
"grad_norm": 1.9472545385360718,
|
| 558 |
+
"learning_rate": 4.228658096201069e-06,
|
| 559 |
+
"loss": 0.1155,
|
| 560 |
+
"step": 35500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.109979633401222,
|
| 564 |
+
"grad_norm": 0.198283389210701,
|
| 565 |
+
"learning_rate": 4.138379315325726e-06,
|
| 566 |
+
"loss": 0.0849,
|
| 567 |
+
"step": 36000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.194840461642905,
|
| 571 |
+
"grad_norm": 0.04895230382680893,
|
| 572 |
+
"learning_rate": 4.048100534450383e-06,
|
| 573 |
+
"loss": 0.0893,
|
| 574 |
+
"step": 36500
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 6.279701289884589,
|
| 578 |
+
"grad_norm": 2.117960214614868,
|
| 579 |
+
"learning_rate": 3.9578217535750406e-06,
|
| 580 |
+
"loss": 0.0888,
|
| 581 |
+
"step": 37000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 6.364562118126273,
|
| 585 |
+
"grad_norm": 57.07741165161133,
|
| 586 |
+
"learning_rate": 3.867542972699697e-06,
|
| 587 |
+
"loss": 0.094,
|
| 588 |
+
"step": 37500
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 6.449422946367957,
|
| 592 |
+
"grad_norm": 0.016213122755289078,
|
| 593 |
+
"learning_rate": 3.777264191824354e-06,
|
| 594 |
+
"loss": 0.0814,
|
| 595 |
+
"step": 38000
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.53428377460964,
|
| 599 |
+
"grad_norm": 70.86405181884766,
|
| 600 |
+
"learning_rate": 3.686985410949011e-06,
|
| 601 |
+
"loss": 0.0783,
|
| 602 |
+
"step": 38500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.619144602851324,
|
| 606 |
+
"grad_norm": 0.039318665862083435,
|
| 607 |
+
"learning_rate": 3.596706630073668e-06,
|
| 608 |
+
"loss": 0.0881,
|
| 609 |
+
"step": 39000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.704005431093007,
|
| 613 |
+
"grad_norm": 0.11302085965871811,
|
| 614 |
+
"learning_rate": 3.5064278491983244e-06,
|
| 615 |
+
"loss": 0.1023,
|
| 616 |
+
"step": 39500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.788866259334691,
|
| 620 |
+
"grad_norm": 0.1393439918756485,
|
| 621 |
+
"learning_rate": 3.4161490683229816e-06,
|
| 622 |
+
"loss": 0.0991,
|
| 623 |
+
"step": 40000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 6.873727087576375,
|
| 627 |
+
"grad_norm": 34.55154800415039,
|
| 628 |
+
"learning_rate": 3.3258702874476384e-06,
|
| 629 |
+
"loss": 0.0921,
|
| 630 |
+
"step": 40500
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.958587915818058,
|
| 634 |
+
"grad_norm": 93.2377700805664,
|
| 635 |
+
"learning_rate": 3.2355915065722956e-06,
|
| 636 |
+
"loss": 0.0909,
|
| 637 |
+
"step": 41000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 7.0,
|
| 641 |
+
"eval_accuracy": 0.9125536992840095,
|
| 642 |
+
"eval_loss": 0.5368871092796326,
|
| 643 |
+
"eval_runtime": 7.6028,
|
| 644 |
+
"eval_samples_per_second": 1377.782,
|
| 645 |
+
"eval_steps_per_second": 86.152,
|
| 646 |
+
"step": 41244
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.043448744059742,
|
| 650 |
+
"grad_norm": 1.839195728302002,
|
| 651 |
+
"learning_rate": 3.1453127256969523e-06,
|
| 652 |
+
"loss": 0.0637,
|
| 653 |
+
"step": 41500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.128309572301426,
|
| 657 |
+
"grad_norm": 0.01967424899339676,
|
| 658 |
+
"learning_rate": 3.055033944821609e-06,
|
| 659 |
+
"loss": 0.0677,
|
| 660 |
+
"step": 42000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.21317040054311,
|
| 664 |
+
"grad_norm": 0.3844837248325348,
|
| 665 |
+
"learning_rate": 2.9647551639462663e-06,
|
| 666 |
+
"loss": 0.0759,
|
| 667 |
+
"step": 42500
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.298031228784793,
|
| 671 |
+
"grad_norm": 0.03412732854485512,
|
| 672 |
+
"learning_rate": 2.874476383070923e-06,
|
| 673 |
+
"loss": 0.0626,
|
| 674 |
+
"step": 43000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.382892057026477,
|
| 678 |
+
"grad_norm": 0.03589722141623497,
|
| 679 |
+
"learning_rate": 2.7841976021955803e-06,
|
| 680 |
+
"loss": 0.0659,
|
| 681 |
+
"step": 43500
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.46775288526816,
|
| 685 |
+
"grad_norm": 0.18037886917591095,
|
| 686 |
+
"learning_rate": 2.693918821320237e-06,
|
| 687 |
+
"loss": 0.0755,
|
| 688 |
+
"step": 44000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.552613713509844,
|
| 692 |
+
"grad_norm": 0.011806238442659378,
|
| 693 |
+
"learning_rate": 2.6036400404448942e-06,
|
| 694 |
+
"loss": 0.072,
|
| 695 |
+
"step": 44500
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.637474541751527,
|
| 699 |
+
"grad_norm": 0.03406194970011711,
|
| 700 |
+
"learning_rate": 2.513361259569551e-06,
|
| 701 |
+
"loss": 0.0724,
|
| 702 |
+
"step": 45000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 7.722335369993211,
|
| 706 |
+
"grad_norm": 0.15559689700603485,
|
| 707 |
+
"learning_rate": 2.4230824786942078e-06,
|
| 708 |
+
"loss": 0.0606,
|
| 709 |
+
"step": 45500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 7.8071961982348945,
|
| 713 |
+
"grad_norm": 209.84703063964844,
|
| 714 |
+
"learning_rate": 2.332803697818865e-06,
|
| 715 |
+
"loss": 0.0661,
|
| 716 |
+
"step": 46000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 7.892057026476579,
|
| 720 |
+
"grad_norm": 0.30213502049446106,
|
| 721 |
+
"learning_rate": 2.2425249169435217e-06,
|
| 722 |
+
"loss": 0.0664,
|
| 723 |
+
"step": 46500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.976917854718262,
|
| 727 |
+
"grad_norm": 34.99961471557617,
|
| 728 |
+
"learning_rate": 2.152246136068179e-06,
|
| 729 |
+
"loss": 0.0687,
|
| 730 |
+
"step": 47000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 8.0,
|
| 734 |
+
"eval_accuracy": 0.9131264916467781,
|
| 735 |
+
"eval_loss": 0.5335070490837097,
|
| 736 |
+
"eval_runtime": 7.6435,
|
| 737 |
+
"eval_samples_per_second": 1370.444,
|
| 738 |
+
"eval_steps_per_second": 85.694,
|
| 739 |
+
"step": 47136
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 8.061778682959945,
|
| 743 |
+
"grad_norm": 0.04247099533677101,
|
| 744 |
+
"learning_rate": 2.0619673551928357e-06,
|
| 745 |
+
"loss": 0.0574,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 8.146639511201629,
|
| 750 |
+
"grad_norm": 36.72422790527344,
|
| 751 |
+
"learning_rate": 1.9716885743174925e-06,
|
| 752 |
+
"loss": 0.0526,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.231500339443313,
|
| 757 |
+
"grad_norm": 0.026284432038664818,
|
| 758 |
+
"learning_rate": 1.8814097934421497e-06,
|
| 759 |
+
"loss": 0.0623,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 8.316361167684997,
|
| 764 |
+
"grad_norm": 0.018155470490455627,
|
| 765 |
+
"learning_rate": 1.7911310125668062e-06,
|
| 766 |
+
"loss": 0.0541,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.40122199592668,
|
| 771 |
+
"grad_norm": 143.7437286376953,
|
| 772 |
+
"learning_rate": 1.7008522316914632e-06,
|
| 773 |
+
"loss": 0.0642,
|
| 774 |
+
"step": 49500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 8.486082824168363,
|
| 778 |
+
"grad_norm": 0.045003194361925125,
|
| 779 |
+
"learning_rate": 1.6105734508161202e-06,
|
| 780 |
+
"loss": 0.0593,
|
| 781 |
+
"step": 50000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 8.570943652410048,
|
| 785 |
+
"grad_norm": 0.1360011100769043,
|
| 786 |
+
"learning_rate": 1.5202946699407772e-06,
|
| 787 |
+
"loss": 0.0539,
|
| 788 |
+
"step": 50500
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 8.655804480651732,
|
| 792 |
+
"grad_norm": 0.14555826783180237,
|
| 793 |
+
"learning_rate": 1.4300158890654341e-06,
|
| 794 |
+
"loss": 0.0399,
|
| 795 |
+
"step": 51000
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.740665308893416,
|
| 799 |
+
"grad_norm": 0.04901694133877754,
|
| 800 |
+
"learning_rate": 1.3397371081900911e-06,
|
| 801 |
+
"loss": 0.0576,
|
| 802 |
+
"step": 51500
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.825526137135098,
|
| 806 |
+
"grad_norm": 0.1550481766462326,
|
| 807 |
+
"learning_rate": 1.249458327314748e-06,
|
| 808 |
+
"loss": 0.0548,
|
| 809 |
+
"step": 52000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.910386965376782,
|
| 813 |
+
"grad_norm": 0.3641352653503418,
|
| 814 |
+
"learning_rate": 1.1591795464394049e-06,
|
| 815 |
+
"loss": 0.048,
|
| 816 |
+
"step": 52500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 8.995247793618466,
|
| 820 |
+
"grad_norm": 112.50800323486328,
|
| 821 |
+
"learning_rate": 1.0689007655640619e-06,
|
| 822 |
+
"loss": 0.0492,
|
| 823 |
+
"step": 53000
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 9.0,
|
| 827 |
+
"eval_accuracy": 0.9149403341288783,
|
| 828 |
+
"eval_loss": 0.5864603519439697,
|
| 829 |
+
"eval_runtime": 7.5667,
|
| 830 |
+
"eval_samples_per_second": 1384.359,
|
| 831 |
+
"eval_steps_per_second": 86.564,
|
| 832 |
+
"step": 53028
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 9.08010862186015,
|
| 836 |
+
"grad_norm": 0.3624100089073181,
|
| 837 |
+
"learning_rate": 9.786219846887188e-07,
|
| 838 |
+
"loss": 0.0463,
|
| 839 |
+
"step": 53500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.164969450101832,
|
| 843 |
+
"grad_norm": 0.07612378150224686,
|
| 844 |
+
"learning_rate": 8.883432038133757e-07,
|
| 845 |
+
"loss": 0.0457,
|
| 846 |
+
"step": 54000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 9.249830278343516,
|
| 850 |
+
"grad_norm": 0.012948695570230484,
|
| 851 |
+
"learning_rate": 7.980644229380327e-07,
|
| 852 |
+
"loss": 0.0353,
|
| 853 |
+
"step": 54500
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 9.3346911065852,
|
| 857 |
+
"grad_norm": 0.05359469726681709,
|
| 858 |
+
"learning_rate": 7.077856420626897e-07,
|
| 859 |
+
"loss": 0.0482,
|
| 860 |
+
"step": 55000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 9.419551934826885,
|
| 864 |
+
"grad_norm": 0.009526051580905914,
|
| 865 |
+
"learning_rate": 6.175068611873466e-07,
|
| 866 |
+
"loss": 0.0361,
|
| 867 |
+
"step": 55500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 9.504412763068567,
|
| 871 |
+
"grad_norm": 0.017915330827236176,
|
| 872 |
+
"learning_rate": 5.272280803120034e-07,
|
| 873 |
+
"loss": 0.0435,
|
| 874 |
+
"step": 56000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 9.589273591310251,
|
| 878 |
+
"grad_norm": 0.00997795257717371,
|
| 879 |
+
"learning_rate": 4.3694929943666047e-07,
|
| 880 |
+
"loss": 0.0416,
|
| 881 |
+
"step": 56500
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 9.674134419551935,
|
| 885 |
+
"grad_norm": 0.008296828716993332,
|
| 886 |
+
"learning_rate": 3.4667051856131735e-07,
|
| 887 |
+
"loss": 0.0447,
|
| 888 |
+
"step": 57000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 9.758995247793619,
|
| 892 |
+
"grad_norm": 0.052807215601205826,
|
| 893 |
+
"learning_rate": 2.5639173768597433e-07,
|
| 894 |
+
"loss": 0.0447,
|
| 895 |
+
"step": 57500
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 9.843856076035301,
|
| 899 |
+
"grad_norm": 0.009740478359162807,
|
| 900 |
+
"learning_rate": 1.6611295681063126e-07,
|
| 901 |
+
"loss": 0.0363,
|
| 902 |
+
"step": 58000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.928716904276985,
|
| 906 |
+
"grad_norm": 0.03634992986917496,
|
| 907 |
+
"learning_rate": 7.583417593528817e-08,
|
| 908 |
+
"loss": 0.04,
|
| 909 |
+
"step": 58500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 10.0,
|
| 913 |
+
"eval_accuracy": 0.9155131264916467,
|
| 914 |
+
"eval_loss": 0.5974043607711792,
|
| 915 |
+
"eval_runtime": 7.5764,
|
| 916 |
+
"eval_samples_per_second": 1382.574,
|
| 917 |
+
"eval_steps_per_second": 86.452,
|
| 918 |
+
"step": 58920
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 10.0,
|
| 922 |
+
"step": 58920,
|
| 923 |
+
"total_flos": 4.325093116880976e+16,
|
| 924 |
+
"train_loss": 0.15497403152929384,
|
| 925 |
+
"train_runtime": 3153.0211,
|
| 926 |
+
"train_samples_per_second": 298.977,
|
| 927 |
+
"train_steps_per_second": 18.687
|
| 928 |
+
}
|
| 929 |
+
],
|
| 930 |
+
"logging_steps": 500,
|
| 931 |
+
"max_steps": 58920,
|
| 932 |
+
"num_input_tokens_seen": 0,
|
| 933 |
+
"num_train_epochs": 10,
|
| 934 |
+
"save_steps": 500,
|
| 935 |
+
"stateful_callbacks": {
|
| 936 |
+
"TrainerControl": {
|
| 937 |
+
"args": {
|
| 938 |
+
"should_epoch_stop": false,
|
| 939 |
+
"should_evaluate": false,
|
| 940 |
+
"should_log": false,
|
| 941 |
+
"should_save": true,
|
| 942 |
+
"should_training_stop": true
|
| 943 |
+
},
|
| 944 |
+
"attributes": {}
|
| 945 |
+
}
|
| 946 |
+
},
|
| 947 |
+
"total_flos": 4.325093116880976e+16,
|
| 948 |
+
"train_batch_size": 16,
|
| 949 |
+
"trial_name": null,
|
| 950 |
+
"trial_params": null
|
| 951 |
+
}
|
qnli/roberta-base_lr1e-05/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:163a2c893191bc7b85f82770dff9e98820bf803b5f40dd1185d3248be6f71ef7
|
| 3 |
+
size 5240
|
qnli/roberta-base_lr1e-05/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-large_lr1e-05/classifier_head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:910f89cb41bf3b0a5b885286bdb2755d9d6f3ac43c40fd082bac7a4322376f0b
|
| 3 |
+
size 4210000
|
qnli/roberta-large_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/projects/shikexuan/nlu_model/roberta-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.44.1",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
qnli/roberta-large_lr1e-05/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-large_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81c063b3c13a18889d939bb1b247f3d095b2bff777d3870db391909583eaee34
|
| 3 |
+
size 1421495416
|
qnli/roberta-large_lr1e-05/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
qnli/roberta-large_lr1e-05/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qnli/roberta-large_lr1e-05/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": true,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"mask_token": "<mask>",
|
| 51 |
+
"model_max_length": 512,
|
| 52 |
+
"pad_token": "<pad>",
|
| 53 |
+
"sep_token": "</s>",
|
| 54 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 55 |
+
"trim_offsets": true,
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
qnli/roberta-large_lr1e-05/trainer_state.json
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9351789976133652,
|
| 3 |
+
"best_model_checkpoint": "./nlu_finetuned_models/qnli/roberta-large_lr1e-05/checkpoint-58920",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 58920,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08486082824168364,
|
| 13 |
+
"grad_norm": 16.678653717041016,
|
| 14 |
+
"learning_rate": 1.4140271493212672e-06,
|
| 15 |
+
"loss": 0.6978,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1697216564833673,
|
| 20 |
+
"grad_norm": 63.95085525512695,
|
| 21 |
+
"learning_rate": 2.8280542986425343e-06,
|
| 22 |
+
"loss": 0.4843,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.2545824847250509,
|
| 27 |
+
"grad_norm": 32.92603302001953,
|
| 28 |
+
"learning_rate": 4.242081447963801e-06,
|
| 29 |
+
"loss": 0.3731,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3394433129667346,
|
| 34 |
+
"grad_norm": 19.359167098999023,
|
| 35 |
+
"learning_rate": 5.656108597285069e-06,
|
| 36 |
+
"loss": 0.3188,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4243041412084182,
|
| 41 |
+
"grad_norm": 33.35426712036133,
|
| 42 |
+
"learning_rate": 7.070135746606336e-06,
|
| 43 |
+
"loss": 0.3128,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5091649694501018,
|
| 48 |
+
"grad_norm": 21.905170440673828,
|
| 49 |
+
"learning_rate": 8.484162895927603e-06,
|
| 50 |
+
"loss": 0.2933,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.5940257976917854,
|
| 55 |
+
"grad_norm": 9.101493835449219,
|
| 56 |
+
"learning_rate": 9.89819004524887e-06,
|
| 57 |
+
"loss": 0.2876,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6788866259334692,
|
| 62 |
+
"grad_norm": 23.2750301361084,
|
| 63 |
+
"learning_rate": 9.916221291347683e-06,
|
| 64 |
+
"loss": 0.2731,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.7637474541751528,
|
| 69 |
+
"grad_norm": 28.954246520996094,
|
| 70 |
+
"learning_rate": 9.825942510472339e-06,
|
| 71 |
+
"loss": 0.2744,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.8486082824168364,
|
| 76 |
+
"grad_norm": 6.048152923583984,
|
| 77 |
+
"learning_rate": 9.735663729596995e-06,
|
| 78 |
+
"loss": 0.2572,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.93346911065852,
|
| 83 |
+
"grad_norm": 24.859359741210938,
|
| 84 |
+
"learning_rate": 9.645384948721653e-06,
|
| 85 |
+
"loss": 0.2417,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.0,
|
| 90 |
+
"eval_accuracy": 0.918854415274463,
|
| 91 |
+
"eval_loss": 0.21485264599323273,
|
| 92 |
+
"eval_runtime": 20.1735,
|
| 93 |
+
"eval_samples_per_second": 519.244,
|
| 94 |
+
"eval_steps_per_second": 32.468,
|
| 95 |
+
"step": 5892
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.0183299389002036,
|
| 99 |
+
"grad_norm": 16.01885986328125,
|
| 100 |
+
"learning_rate": 9.55510616784631e-06,
|
| 101 |
+
"loss": 0.2369,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1031907671418872,
|
| 106 |
+
"grad_norm": 17.05298614501953,
|
| 107 |
+
"learning_rate": 9.464827386970968e-06,
|
| 108 |
+
"loss": 0.2333,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1880515953835709,
|
| 113 |
+
"grad_norm": 11.151917457580566,
|
| 114 |
+
"learning_rate": 9.374548606095624e-06,
|
| 115 |
+
"loss": 0.2257,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.2729124236252547,
|
| 120 |
+
"grad_norm": 12.592936515808105,
|
| 121 |
+
"learning_rate": 9.28426982522028e-06,
|
| 122 |
+
"loss": 0.2207,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.3577732518669383,
|
| 127 |
+
"grad_norm": 28.379621505737305,
|
| 128 |
+
"learning_rate": 9.193991044344937e-06,
|
| 129 |
+
"loss": 0.2167,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.442634080108622,
|
| 134 |
+
"grad_norm": 25.368091583251953,
|
| 135 |
+
"learning_rate": 9.103712263469595e-06,
|
| 136 |
+
"loss": 0.228,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.5274949083503055,
|
| 141 |
+
"grad_norm": 0.8118745684623718,
|
| 142 |
+
"learning_rate": 9.013433482594251e-06,
|
| 143 |
+
"loss": 0.2066,
|
| 144 |
+
"step": 9000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.6123557365919892,
|
| 148 |
+
"grad_norm": 22.968481063842773,
|
| 149 |
+
"learning_rate": 8.92315470171891e-06,
|
| 150 |
+
"loss": 0.2369,
|
| 151 |
+
"step": 9500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.6972165648336728,
|
| 155 |
+
"grad_norm": 3.1241774559020996,
|
| 156 |
+
"learning_rate": 8.832875920843566e-06,
|
| 157 |
+
"loss": 0.2249,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.7820773930753564,
|
| 162 |
+
"grad_norm": 1.0083893537521362,
|
| 163 |
+
"learning_rate": 8.742597139968224e-06,
|
| 164 |
+
"loss": 0.2187,
|
| 165 |
+
"step": 10500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.86693822131704,
|
| 169 |
+
"grad_norm": 1.5048400163650513,
|
| 170 |
+
"learning_rate": 8.65231835909288e-06,
|
| 171 |
+
"loss": 0.2279,
|
| 172 |
+
"step": 11000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.9517990495587236,
|
| 176 |
+
"grad_norm": 3.1371958255767822,
|
| 177 |
+
"learning_rate": 8.562039578217536e-06,
|
| 178 |
+
"loss": 0.2273,
|
| 179 |
+
"step": 11500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_accuracy": 0.9297374701670644,
|
| 184 |
+
"eval_loss": 0.2355683594942093,
|
| 185 |
+
"eval_runtime": 20.1536,
|
| 186 |
+
"eval_samples_per_second": 519.759,
|
| 187 |
+
"eval_steps_per_second": 32.5,
|
| 188 |
+
"step": 11784
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.0366598778004072,
|
| 192 |
+
"grad_norm": 101.15937805175781,
|
| 193 |
+
"learning_rate": 8.471760797342193e-06,
|
| 194 |
+
"loss": 0.1848,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.121520706042091,
|
| 199 |
+
"grad_norm": 0.7154932022094727,
|
| 200 |
+
"learning_rate": 8.381482016466849e-06,
|
| 201 |
+
"loss": 0.1507,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.2063815342837745,
|
| 206 |
+
"grad_norm": 0.0770418643951416,
|
| 207 |
+
"learning_rate": 8.291203235591507e-06,
|
| 208 |
+
"loss": 0.1756,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.291242362525458,
|
| 213 |
+
"grad_norm": 0.3646220862865448,
|
| 214 |
+
"learning_rate": 8.200924454716163e-06,
|
| 215 |
+
"loss": 0.1709,
|
| 216 |
+
"step": 13500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.3761031907671417,
|
| 220 |
+
"grad_norm": 14.423376083374023,
|
| 221 |
+
"learning_rate": 8.110645673840822e-06,
|
| 222 |
+
"loss": 0.1721,
|
| 223 |
+
"step": 14000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.4609640190088253,
|
| 227 |
+
"grad_norm": 0.09149379283189774,
|
| 228 |
+
"learning_rate": 8.020366892965478e-06,
|
| 229 |
+
"loss": 0.1718,
|
| 230 |
+
"step": 14500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.5458248472505094,
|
| 234 |
+
"grad_norm": 3.870497226715088,
|
| 235 |
+
"learning_rate": 7.930088112090136e-06,
|
| 236 |
+
"loss": 0.1886,
|
| 237 |
+
"step": 15000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.6306856754921926,
|
| 241 |
+
"grad_norm": 31.083229064941406,
|
| 242 |
+
"learning_rate": 7.839809331214792e-06,
|
| 243 |
+
"loss": 0.1674,
|
| 244 |
+
"step": 15500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.7155465037338766,
|
| 248 |
+
"grad_norm": 2.2668426036834717,
|
| 249 |
+
"learning_rate": 7.749530550339449e-06,
|
| 250 |
+
"loss": 0.1708,
|
| 251 |
+
"step": 16000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.8004073319755602,
|
| 255 |
+
"grad_norm": 0.7557393908500671,
|
| 256 |
+
"learning_rate": 7.659251769464105e-06,
|
| 257 |
+
"loss": 0.1778,
|
| 258 |
+
"step": 16500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.885268160217244,
|
| 262 |
+
"grad_norm": 8.960997581481934,
|
| 263 |
+
"learning_rate": 7.568972988588763e-06,
|
| 264 |
+
"loss": 0.1731,
|
| 265 |
+
"step": 17000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.9701289884589275,
|
| 269 |
+
"grad_norm": 25.05205726623535,
|
| 270 |
+
"learning_rate": 7.478694207713419e-06,
|
| 271 |
+
"loss": 0.1845,
|
| 272 |
+
"step": 17500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.0,
|
| 276 |
+
"eval_accuracy": 0.9265871121718378,
|
| 277 |
+
"eval_loss": 0.33601415157318115,
|
| 278 |
+
"eval_runtime": 20.1145,
|
| 279 |
+
"eval_samples_per_second": 520.769,
|
| 280 |
+
"eval_steps_per_second": 32.564,
|
| 281 |
+
"step": 17676
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.054989816700611,
|
| 285 |
+
"grad_norm": 0.1671181321144104,
|
| 286 |
+
"learning_rate": 7.388415426838077e-06,
|
| 287 |
+
"loss": 0.13,
|
| 288 |
+
"step": 18000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.1398506449422947,
|
| 292 |
+
"grad_norm": 0.8593884706497192,
|
| 293 |
+
"learning_rate": 7.298136645962734e-06,
|
| 294 |
+
"loss": 0.1053,
|
| 295 |
+
"step": 18500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.2247114731839783,
|
| 299 |
+
"grad_norm": 0.1440092772245407,
|
| 300 |
+
"learning_rate": 7.207857865087391e-06,
|
| 301 |
+
"loss": 0.1287,
|
| 302 |
+
"step": 19000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.309572301425662,
|
| 306 |
+
"grad_norm": 285.67822265625,
|
| 307 |
+
"learning_rate": 7.117579084212047e-06,
|
| 308 |
+
"loss": 0.1142,
|
| 309 |
+
"step": 19500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.3944331296673456,
|
| 313 |
+
"grad_norm": 21.172086715698242,
|
| 314 |
+
"learning_rate": 7.027300303336704e-06,
|
| 315 |
+
"loss": 0.1118,
|
| 316 |
+
"step": 20000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.479293957909029,
|
| 320 |
+
"grad_norm": 0.18984951078891754,
|
| 321 |
+
"learning_rate": 6.937021522461362e-06,
|
| 322 |
+
"loss": 0.1148,
|
| 323 |
+
"step": 20500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.564154786150713,
|
| 327 |
+
"grad_norm": 15.848613739013672,
|
| 328 |
+
"learning_rate": 6.846742741586018e-06,
|
| 329 |
+
"loss": 0.1146,
|
| 330 |
+
"step": 21000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.6490156143923964,
|
| 334 |
+
"grad_norm": 0.1506059467792511,
|
| 335 |
+
"learning_rate": 6.756463960710675e-06,
|
| 336 |
+
"loss": 0.1301,
|
| 337 |
+
"step": 21500
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.73387644263408,
|
| 341 |
+
"grad_norm": 5.585486888885498,
|
| 342 |
+
"learning_rate": 6.6661851798353315e-06,
|
| 343 |
+
"loss": 0.1268,
|
| 344 |
+
"step": 22000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.8187372708757636,
|
| 348 |
+
"grad_norm": 0.15387825667858124,
|
| 349 |
+
"learning_rate": 6.57590639895999e-06,
|
| 350 |
+
"loss": 0.1202,
|
| 351 |
+
"step": 22500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.9035980991174473,
|
| 355 |
+
"grad_norm": 30.81467056274414,
|
| 356 |
+
"learning_rate": 6.485627618084646e-06,
|
| 357 |
+
"loss": 0.1268,
|
| 358 |
+
"step": 23000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.988458927359131,
|
| 362 |
+
"grad_norm": 10.909303665161133,
|
| 363 |
+
"learning_rate": 6.395348837209303e-06,
|
| 364 |
+
"loss": 0.1134,
|
| 365 |
+
"step": 23500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.0,
|
| 369 |
+
"eval_accuracy": 0.9308830548926015,
|
| 370 |
+
"eval_loss": 0.34831878542900085,
|
| 371 |
+
"eval_runtime": 20.2728,
|
| 372 |
+
"eval_samples_per_second": 516.702,
|
| 373 |
+
"eval_steps_per_second": 32.309,
|
| 374 |
+
"step": 23568
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.0733197556008145,
|
| 378 |
+
"grad_norm": 0.025312168523669243,
|
| 379 |
+
"learning_rate": 6.3050700563339595e-06,
|
| 380 |
+
"loss": 0.0743,
|
| 381 |
+
"step": 24000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.158180583842499,
|
| 385 |
+
"grad_norm": 0.053047724068164825,
|
| 386 |
+
"learning_rate": 6.214791275458617e-06,
|
| 387 |
+
"loss": 0.0624,
|
| 388 |
+
"step": 24500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.243041412084182,
|
| 392 |
+
"grad_norm": 0.04022592678666115,
|
| 393 |
+
"learning_rate": 6.124512494583274e-06,
|
| 394 |
+
"loss": 0.0802,
|
| 395 |
+
"step": 25000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.327902240325866,
|
| 399 |
+
"grad_norm": 7.32888126373291,
|
| 400 |
+
"learning_rate": 6.034233713707931e-06,
|
| 401 |
+
"loss": 0.0784,
|
| 402 |
+
"step": 25500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.412763068567549,
|
| 406 |
+
"grad_norm": 50.06730270385742,
|
| 407 |
+
"learning_rate": 5.943954932832587e-06,
|
| 408 |
+
"loss": 0.0886,
|
| 409 |
+
"step": 26000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.497623896809233,
|
| 413 |
+
"grad_norm": 0.16255952417850494,
|
| 414 |
+
"learning_rate": 5.853676151957245e-06,
|
| 415 |
+
"loss": 0.0796,
|
| 416 |
+
"step": 26500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.582484725050916,
|
| 420 |
+
"grad_norm": 18.90667152404785,
|
| 421 |
+
"learning_rate": 5.763397371081901e-06,
|
| 422 |
+
"loss": 0.0828,
|
| 423 |
+
"step": 27000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.6673455532926,
|
| 427 |
+
"grad_norm": 0.08691630512475967,
|
| 428 |
+
"learning_rate": 5.673118590206559e-06,
|
| 429 |
+
"loss": 0.0949,
|
| 430 |
+
"step": 27500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.752206381534283,
|
| 434 |
+
"grad_norm": 0.023470040410757065,
|
| 435 |
+
"learning_rate": 5.582839809331215e-06,
|
| 436 |
+
"loss": 0.0834,
|
| 437 |
+
"step": 28000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.8370672097759675,
|
| 441 |
+
"grad_norm": 45.75636291503906,
|
| 442 |
+
"learning_rate": 5.492561028455872e-06,
|
| 443 |
+
"loss": 0.0849,
|
| 444 |
+
"step": 28500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.921928038017651,
|
| 448 |
+
"grad_norm": 0.02307937666773796,
|
| 449 |
+
"learning_rate": 5.402282247580529e-06,
|
| 450 |
+
"loss": 0.0814,
|
| 451 |
+
"step": 29000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 5.0,
|
| 455 |
+
"eval_accuracy": 0.9324105011933175,
|
| 456 |
+
"eval_loss": 0.4098624885082245,
|
| 457 |
+
"eval_runtime": 20.1611,
|
| 458 |
+
"eval_samples_per_second": 519.565,
|
| 459 |
+
"eval_steps_per_second": 32.488,
|
| 460 |
+
"step": 29460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.006788866259335,
|
| 464 |
+
"grad_norm": 0.1366582214832306,
|
| 465 |
+
"learning_rate": 5.312003466705185e-06,
|
| 466 |
+
"loss": 0.0852,
|
| 467 |
+
"step": 29500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.091649694501018,
|
| 471 |
+
"grad_norm": 4.484254360198975,
|
| 472 |
+
"learning_rate": 5.221724685829843e-06,
|
| 473 |
+
"loss": 0.043,
|
| 474 |
+
"step": 30000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.176510522742702,
|
| 478 |
+
"grad_norm": 0.023312659934163094,
|
| 479 |
+
"learning_rate": 5.1314459049545e-06,
|
| 480 |
+
"loss": 0.0561,
|
| 481 |
+
"step": 30500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.261371350984386,
|
| 485 |
+
"grad_norm": 0.21991141140460968,
|
| 486 |
+
"learning_rate": 5.041167124079157e-06,
|
| 487 |
+
"loss": 0.0571,
|
| 488 |
+
"step": 31000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.346232179226069,
|
| 492 |
+
"grad_norm": 0.059541311115026474,
|
| 493 |
+
"learning_rate": 4.950888343203813e-06,
|
| 494 |
+
"loss": 0.057,
|
| 495 |
+
"step": 31500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.431093007467753,
|
| 499 |
+
"grad_norm": 0.008009650744497776,
|
| 500 |
+
"learning_rate": 4.86060956232847e-06,
|
| 501 |
+
"loss": 0.0581,
|
| 502 |
+
"step": 32000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.515953835709436,
|
| 506 |
+
"grad_norm": 22.845497131347656,
|
| 507 |
+
"learning_rate": 4.7703307814531275e-06,
|
| 508 |
+
"loss": 0.0528,
|
| 509 |
+
"step": 32500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.6008146639511205,
|
| 513 |
+
"grad_norm": 0.04798487201333046,
|
| 514 |
+
"learning_rate": 4.680052000577785e-06,
|
| 515 |
+
"loss": 0.0605,
|
| 516 |
+
"step": 33000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 5.685675492192804,
|
| 520 |
+
"grad_norm": 0.03193668648600578,
|
| 521 |
+
"learning_rate": 4.589773219702441e-06,
|
| 522 |
+
"loss": 0.0411,
|
| 523 |
+
"step": 33500
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 5.770536320434488,
|
| 527 |
+
"grad_norm": 0.0065535930916666985,
|
| 528 |
+
"learning_rate": 4.499494438827098e-06,
|
| 529 |
+
"loss": 0.0505,
|
| 530 |
+
"step": 34000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.855397148676171,
|
| 534 |
+
"grad_norm": 0.14513394236564636,
|
| 535 |
+
"learning_rate": 4.4092156579517554e-06,
|
| 536 |
+
"loss": 0.0656,
|
| 537 |
+
"step": 34500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.940257976917855,
|
| 541 |
+
"grad_norm": 0.01327830646187067,
|
| 542 |
+
"learning_rate": 4.318936877076413e-06,
|
| 543 |
+
"loss": 0.0608,
|
| 544 |
+
"step": 35000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 6.0,
|
| 548 |
+
"eval_accuracy": 0.9305966587112172,
|
| 549 |
+
"eval_loss": 0.4498295783996582,
|
| 550 |
+
"eval_runtime": 20.1533,
|
| 551 |
+
"eval_samples_per_second": 519.766,
|
| 552 |
+
"eval_steps_per_second": 32.501,
|
| 553 |
+
"step": 35352
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.025118805159538,
|
| 557 |
+
"grad_norm": 0.008021282963454723,
|
| 558 |
+
"learning_rate": 4.228658096201069e-06,
|
| 559 |
+
"loss": 0.0506,
|
| 560 |
+
"step": 35500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.109979633401222,
|
| 564 |
+
"grad_norm": 0.02884720452129841,
|
| 565 |
+
"learning_rate": 4.138379315325726e-06,
|
| 566 |
+
"loss": 0.0373,
|
| 567 |
+
"step": 36000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.194840461642905,
|
| 571 |
+
"grad_norm": 0.05565548315644264,
|
| 572 |
+
"learning_rate": 4.048100534450383e-06,
|
| 573 |
+
"loss": 0.0275,
|
| 574 |
+
"step": 36500
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 6.279701289884589,
|
| 578 |
+
"grad_norm": 0.004155606031417847,
|
| 579 |
+
"learning_rate": 3.9578217535750406e-06,
|
| 580 |
+
"loss": 0.0363,
|
| 581 |
+
"step": 37000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 6.364562118126273,
|
| 585 |
+
"grad_norm": 0.012492923066020012,
|
| 586 |
+
"learning_rate": 3.867542972699697e-06,
|
| 587 |
+
"loss": 0.0279,
|
| 588 |
+
"step": 37500
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 6.449422946367957,
|
| 592 |
+
"grad_norm": 0.0035774342250078917,
|
| 593 |
+
"learning_rate": 3.777264191824354e-06,
|
| 594 |
+
"loss": 0.0336,
|
| 595 |
+
"step": 38000
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.53428377460964,
|
| 599 |
+
"grad_norm": 0.029567407444119453,
|
| 600 |
+
"learning_rate": 3.686985410949011e-06,
|
| 601 |
+
"loss": 0.0401,
|
| 602 |
+
"step": 38500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.619144602851324,
|
| 606 |
+
"grad_norm": 0.0071405465714633465,
|
| 607 |
+
"learning_rate": 3.596706630073668e-06,
|
| 608 |
+
"loss": 0.0314,
|
| 609 |
+
"step": 39000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.704005431093007,
|
| 613 |
+
"grad_norm": 0.02880735881626606,
|
| 614 |
+
"learning_rate": 3.5064278491983244e-06,
|
| 615 |
+
"loss": 0.0402,
|
| 616 |
+
"step": 39500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.788866259334691,
|
| 620 |
+
"grad_norm": 0.002806098898872733,
|
| 621 |
+
"learning_rate": 3.4161490683229816e-06,
|
| 622 |
+
"loss": 0.041,
|
| 623 |
+
"step": 40000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 6.873727087576375,
|
| 627 |
+
"grad_norm": 0.005201249849051237,
|
| 628 |
+
"learning_rate": 3.3258702874476384e-06,
|
| 629 |
+
"loss": 0.0299,
|
| 630 |
+
"step": 40500
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.958587915818058,
|
| 634 |
+
"grad_norm": 31.786758422851562,
|
| 635 |
+
"learning_rate": 3.2355915065722956e-06,
|
| 636 |
+
"loss": 0.037,
|
| 637 |
+
"step": 41000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 7.0,
|
| 641 |
+
"eval_accuracy": 0.9321241050119332,
|
| 642 |
+
"eval_loss": 0.5288263559341431,
|
| 643 |
+
"eval_runtime": 20.1543,
|
| 644 |
+
"eval_samples_per_second": 519.74,
|
| 645 |
+
"eval_steps_per_second": 32.499,
|
| 646 |
+
"step": 41244
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.043448744059742,
|
| 650 |
+
"grad_norm": 0.02835765853524208,
|
| 651 |
+
"learning_rate": 3.1453127256969523e-06,
|
| 652 |
+
"loss": 0.0321,
|
| 653 |
+
"step": 41500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.128309572301426,
|
| 657 |
+
"grad_norm": 0.0013393410481512547,
|
| 658 |
+
"learning_rate": 3.055033944821609e-06,
|
| 659 |
+
"loss": 0.0175,
|
| 660 |
+
"step": 42000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.21317040054311,
|
| 664 |
+
"grad_norm": 0.012179546058177948,
|
| 665 |
+
"learning_rate": 2.9647551639462663e-06,
|
| 666 |
+
"loss": 0.0219,
|
| 667 |
+
"step": 42500
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.298031228784793,
|
| 671 |
+
"grad_norm": 228.60903930664062,
|
| 672 |
+
"learning_rate": 2.874476383070923e-06,
|
| 673 |
+
"loss": 0.0176,
|
| 674 |
+
"step": 43000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.382892057026477,
|
| 678 |
+
"grad_norm": 0.0011996960965916514,
|
| 679 |
+
"learning_rate": 2.7841976021955803e-06,
|
| 680 |
+
"loss": 0.0295,
|
| 681 |
+
"step": 43500
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.46775288526816,
|
| 685 |
+
"grad_norm": 0.0019071695860475302,
|
| 686 |
+
"learning_rate": 2.693918821320237e-06,
|
| 687 |
+
"loss": 0.0218,
|
| 688 |
+
"step": 44000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.552613713509844,
|
| 692 |
+
"grad_norm": 0.01705249398946762,
|
| 693 |
+
"learning_rate": 2.6036400404448942e-06,
|
| 694 |
+
"loss": 0.0237,
|
| 695 |
+
"step": 44500
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.637474541751527,
|
| 699 |
+
"grad_norm": 0.0021223679650574923,
|
| 700 |
+
"learning_rate": 2.513361259569551e-06,
|
| 701 |
+
"loss": 0.0216,
|
| 702 |
+
"step": 45000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 7.722335369993211,
|
| 706 |
+
"grad_norm": 0.0015007400652393699,
|
| 707 |
+
"learning_rate": 2.4230824786942078e-06,
|
| 708 |
+
"loss": 0.0178,
|
| 709 |
+
"step": 45500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 7.8071961982348945,
|
| 713 |
+
"grad_norm": 0.0023892533499747515,
|
| 714 |
+
"learning_rate": 2.332803697818865e-06,
|
| 715 |
+
"loss": 0.0315,
|
| 716 |
+
"step": 46000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 7.892057026476579,
|
| 720 |
+
"grad_norm": 0.14909300208091736,
|
| 721 |
+
"learning_rate": 2.2425249169435217e-06,
|
| 722 |
+
"loss": 0.016,
|
| 723 |
+
"step": 46500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.976917854718262,
|
| 727 |
+
"grad_norm": 0.0012619789922609925,
|
| 728 |
+
"learning_rate": 2.152246136068179e-06,
|
| 729 |
+
"loss": 0.0211,
|
| 730 |
+
"step": 47000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 8.0,
|
| 734 |
+
"eval_accuracy": 0.9331742243436754,
|
| 735 |
+
"eval_loss": 0.5570583343505859,
|
| 736 |
+
"eval_runtime": 20.2838,
|
| 737 |
+
"eval_samples_per_second": 516.421,
|
| 738 |
+
"eval_steps_per_second": 32.292,
|
| 739 |
+
"step": 47136
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 8.061778682959945,
|
| 743 |
+
"grad_norm": 0.0995703861117363,
|
| 744 |
+
"learning_rate": 2.0619673551928357e-06,
|
| 745 |
+
"loss": 0.0199,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 8.146639511201629,
|
| 750 |
+
"grad_norm": 0.4637010097503662,
|
| 751 |
+
"learning_rate": 1.9716885743174925e-06,
|
| 752 |
+
"loss": 0.0082,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.231500339443313,
|
| 757 |
+
"grad_norm": 0.002060811733826995,
|
| 758 |
+
"learning_rate": 1.8814097934421497e-06,
|
| 759 |
+
"loss": 0.0114,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 8.316361167684997,
|
| 764 |
+
"grad_norm": 0.0012016426771879196,
|
| 765 |
+
"learning_rate": 1.7911310125668062e-06,
|
| 766 |
+
"loss": 0.0124,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.40122199592668,
|
| 771 |
+
"grad_norm": 0.0012142487103119493,
|
| 772 |
+
"learning_rate": 1.7008522316914632e-06,
|
| 773 |
+
"loss": 0.0148,
|
| 774 |
+
"step": 49500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 8.486082824168363,
|
| 778 |
+
"grad_norm": 0.0008126778993755579,
|
| 779 |
+
"learning_rate": 1.6105734508161202e-06,
|
| 780 |
+
"loss": 0.0099,
|
| 781 |
+
"step": 50000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 8.570943652410048,
|
| 785 |
+
"grad_norm": 0.00788600742816925,
|
| 786 |
+
"learning_rate": 1.5202946699407772e-06,
|
| 787 |
+
"loss": 0.018,
|
| 788 |
+
"step": 50500
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 8.655804480651732,
|
| 792 |
+
"grad_norm": 0.4367898404598236,
|
| 793 |
+
"learning_rate": 1.4300158890654341e-06,
|
| 794 |
+
"loss": 0.0098,
|
| 795 |
+
"step": 51000
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.740665308893416,
|
| 799 |
+
"grad_norm": 0.001816450385376811,
|
| 800 |
+
"learning_rate": 1.3397371081900911e-06,
|
| 801 |
+
"loss": 0.0133,
|
| 802 |
+
"step": 51500
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.825526137135098,
|
| 806 |
+
"grad_norm": 0.08108772337436676,
|
| 807 |
+
"learning_rate": 1.249458327314748e-06,
|
| 808 |
+
"loss": 0.0188,
|
| 809 |
+
"step": 52000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.910386965376782,
|
| 813 |
+
"grad_norm": 0.037823718041181564,
|
| 814 |
+
"learning_rate": 1.1591795464394049e-06,
|
| 815 |
+
"loss": 0.0152,
|
| 816 |
+
"step": 52500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 8.995247793618466,
|
| 820 |
+
"grad_norm": 0.01213313639163971,
|
| 821 |
+
"learning_rate": 1.0689007655640619e-06,
|
| 822 |
+
"loss": 0.0126,
|
| 823 |
+
"step": 53000
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 9.0,
|
| 827 |
+
"eval_accuracy": 0.9338424821002387,
|
| 828 |
+
"eval_loss": 0.5792200565338135,
|
| 829 |
+
"eval_runtime": 20.1485,
|
| 830 |
+
"eval_samples_per_second": 519.889,
|
| 831 |
+
"eval_steps_per_second": 32.509,
|
| 832 |
+
"step": 53028
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 9.08010862186015,
|
| 836 |
+
"grad_norm": 0.0018375592771917582,
|
| 837 |
+
"learning_rate": 9.786219846887188e-07,
|
| 838 |
+
"loss": 0.0099,
|
| 839 |
+
"step": 53500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.164969450101832,
|
| 843 |
+
"grad_norm": 0.07955571264028549,
|
| 844 |
+
"learning_rate": 8.883432038133757e-07,
|
| 845 |
+
"loss": 0.0088,
|
| 846 |
+
"step": 54000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 9.249830278343516,
|
| 850 |
+
"grad_norm": 0.005015624221414328,
|
| 851 |
+
"learning_rate": 7.980644229380327e-07,
|
| 852 |
+
"loss": 0.0078,
|
| 853 |
+
"step": 54500
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 9.3346911065852,
|
| 857 |
+
"grad_norm": 0.006615865975618362,
|
| 858 |
+
"learning_rate": 7.077856420626897e-07,
|
| 859 |
+
"loss": 0.0173,
|
| 860 |
+
"step": 55000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 9.419551934826885,
|
| 864 |
+
"grad_norm": 0.004652131348848343,
|
| 865 |
+
"learning_rate": 6.175068611873466e-07,
|
| 866 |
+
"loss": 0.0088,
|
| 867 |
+
"step": 55500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 9.504412763068567,
|
| 871 |
+
"grad_norm": 0.0025109213311225176,
|
| 872 |
+
"learning_rate": 5.272280803120034e-07,
|
| 873 |
+
"loss": 0.0075,
|
| 874 |
+
"step": 56000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 9.589273591310251,
|
| 878 |
+
"grad_norm": 0.412564754486084,
|
| 879 |
+
"learning_rate": 4.3694929943666047e-07,
|
| 880 |
+
"loss": 0.0107,
|
| 881 |
+
"step": 56500
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 9.674134419551935,
|
| 885 |
+
"grad_norm": 0.0012215793831273913,
|
| 886 |
+
"learning_rate": 3.4667051856131735e-07,
|
| 887 |
+
"loss": 0.0063,
|
| 888 |
+
"step": 57000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 9.758995247793619,
|
| 892 |
+
"grad_norm": 0.0011442139511927962,
|
| 893 |
+
"learning_rate": 2.5639173768597433e-07,
|
| 894 |
+
"loss": 0.0067,
|
| 895 |
+
"step": 57500
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 9.843856076035301,
|
| 899 |
+
"grad_norm": 0.000626246037427336,
|
| 900 |
+
"learning_rate": 1.6611295681063126e-07,
|
| 901 |
+
"loss": 0.0092,
|
| 902 |
+
"step": 58000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.928716904276985,
|
| 906 |
+
"grad_norm": 0.0007085532415658236,
|
| 907 |
+
"learning_rate": 7.583417593528817e-08,
|
| 908 |
+
"loss": 0.0017,
|
| 909 |
+
"step": 58500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 10.0,
|
| 913 |
+
"eval_accuracy": 0.9351789976133652,
|
| 914 |
+
"eval_loss": 0.6194970011711121,
|
| 915 |
+
"eval_runtime": 20.0261,
|
| 916 |
+
"eval_samples_per_second": 523.066,
|
| 917 |
+
"eval_steps_per_second": 32.707,
|
| 918 |
+
"step": 58920
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 10.0,
|
| 922 |
+
"step": 58920,
|
| 923 |
+
"total_flos": 1.5319348399938845e+17,
|
| 924 |
+
"train_loss": 0.10688133286988744,
|
| 925 |
+
"train_runtime": 7408.3391,
|
| 926 |
+
"train_samples_per_second": 127.246,
|
| 927 |
+
"train_steps_per_second": 7.953
|
| 928 |
+
}
|
| 929 |
+
],
|
| 930 |
+
"logging_steps": 500,
|
| 931 |
+
"max_steps": 58920,
|
| 932 |
+
"num_input_tokens_seen": 0,
|
| 933 |
+
"num_train_epochs": 10,
|
| 934 |
+
"save_steps": 500,
|
| 935 |
+
"stateful_callbacks": {
|
| 936 |
+
"TrainerControl": {
|
| 937 |
+
"args": {
|
| 938 |
+
"should_epoch_stop": false,
|
| 939 |
+
"should_evaluate": false,
|
| 940 |
+
"should_log": false,
|
| 941 |
+
"should_save": true,
|
| 942 |
+
"should_training_stop": true
|
| 943 |
+
},
|
| 944 |
+
"attributes": {}
|
| 945 |
+
}
|
| 946 |
+
},
|
| 947 |
+
"total_flos": 1.5319348399938845e+17,
|
| 948 |
+
"train_batch_size": 16,
|
| 949 |
+
"trial_name": null,
|
| 950 |
+
"trial_params": null
|
| 951 |
+
}
|
qnli/roberta-large_lr1e-05/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c12712b0b1d09c69529ff19d1c2cd1a316044b8c570dd167f98987eac4c0a61e
|
| 3 |
+
size 5240
|
qnli/roberta-large_lr1e-05/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/bert-base-uncased_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/shikexuan/nlu_model/bert-base-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.45.2",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
qqp/bert-base-uncased_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac9af916f313248f90ccf6280ea58a566b442997b9f30a9be5faf981f5e11621
|
| 3 |
+
size 437958648
|
qqp/bert-base-uncased_lr1e-05/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
qqp/bert-base-uncased_lr1e-05/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/bert-base-uncased_lr1e-05/tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
qqp/bert-base-uncased_lr1e-05/trainer_state.json
ADDED
|
@@ -0,0 +1,3015 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8931284118506033,
|
| 3 |
+
"best_model_checkpoint": "./nlu_finetuned_models/qqp/bert-base-uncased_lr1e-05/checkpoint-163736",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 204670,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.024429569550984513,
|
| 13 |
+
"grad_norm": 1.8005802631378174,
|
| 14 |
+
"learning_rate": 4.0713296962788053e-07,
|
| 15 |
+
"loss": 0.7002,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.048859139101969026,
|
| 20 |
+
"grad_norm": 5.8743462562561035,
|
| 21 |
+
"learning_rate": 8.142659392557611e-07,
|
| 22 |
+
"loss": 0.6418,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07328870865295353,
|
| 27 |
+
"grad_norm": 7.699524879455566,
|
| 28 |
+
"learning_rate": 1.2213989088836414e-06,
|
| 29 |
+
"loss": 0.5324,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.09771827820393805,
|
| 34 |
+
"grad_norm": 18.407461166381836,
|
| 35 |
+
"learning_rate": 1.6285318785115221e-06,
|
| 36 |
+
"loss": 0.4908,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.12214784775492256,
|
| 41 |
+
"grad_norm": 7.416097164154053,
|
| 42 |
+
"learning_rate": 2.0356648481394024e-06,
|
| 43 |
+
"loss": 0.4535,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14657741730590707,
|
| 48 |
+
"grad_norm": 7.809607028961182,
|
| 49 |
+
"learning_rate": 2.442797817767283e-06,
|
| 50 |
+
"loss": 0.4375,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.17100698685689159,
|
| 55 |
+
"grad_norm": 12.742324829101562,
|
| 56 |
+
"learning_rate": 2.8499307873951637e-06,
|
| 57 |
+
"loss": 0.4178,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1954365564078761,
|
| 62 |
+
"grad_norm": 2.653526782989502,
|
| 63 |
+
"learning_rate": 3.2570637570230442e-06,
|
| 64 |
+
"loss": 0.4058,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2198661259588606,
|
| 69 |
+
"grad_norm": 7.899631500244141,
|
| 70 |
+
"learning_rate": 3.6641967266509243e-06,
|
| 71 |
+
"loss": 0.3962,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.24429569550984512,
|
| 76 |
+
"grad_norm": 9.921072006225586,
|
| 77 |
+
"learning_rate": 4.071329696278805e-06,
|
| 78 |
+
"loss": 0.3878,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2687252650608296,
|
| 83 |
+
"grad_norm": 9.545881271362305,
|
| 84 |
+
"learning_rate": 4.478462665906685e-06,
|
| 85 |
+
"loss": 0.3864,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.29315483461181413,
|
| 90 |
+
"grad_norm": 10.387801170349121,
|
| 91 |
+
"learning_rate": 4.885595635534566e-06,
|
| 92 |
+
"loss": 0.3746,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.31758440416279865,
|
| 97 |
+
"grad_norm": 18.420791625976562,
|
| 98 |
+
"learning_rate": 5.292728605162446e-06,
|
| 99 |
+
"loss": 0.3591,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.34201397371378317,
|
| 104 |
+
"grad_norm": 12.266908645629883,
|
| 105 |
+
"learning_rate": 5.6998615747903275e-06,
|
| 106 |
+
"loss": 0.3567,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3664435432647677,
|
| 111 |
+
"grad_norm": 10.194087982177734,
|
| 112 |
+
"learning_rate": 6.106994544418208e-06,
|
| 113 |
+
"loss": 0.3586,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3908731128157522,
|
| 118 |
+
"grad_norm": 8.542230606079102,
|
| 119 |
+
"learning_rate": 6.5141275140460884e-06,
|
| 120 |
+
"loss": 0.3353,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4153026823667367,
|
| 125 |
+
"grad_norm": 14.4937744140625,
|
| 126 |
+
"learning_rate": 6.921260483673968e-06,
|
| 127 |
+
"loss": 0.3413,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4397322519177212,
|
| 132 |
+
"grad_norm": 9.627971649169922,
|
| 133 |
+
"learning_rate": 7.3283934533018485e-06,
|
| 134 |
+
"loss": 0.3451,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4641618214687057,
|
| 139 |
+
"grad_norm": 10.484150886535645,
|
| 140 |
+
"learning_rate": 7.73552642292973e-06,
|
| 141 |
+
"loss": 0.3309,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.48859139101969024,
|
| 146 |
+
"grad_norm": 4.860393047332764,
|
| 147 |
+
"learning_rate": 8.14265939255761e-06,
|
| 148 |
+
"loss": 0.3251,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5130209605706747,
|
| 153 |
+
"grad_norm": 9.575475692749023,
|
| 154 |
+
"learning_rate": 8.54979236218549e-06,
|
| 155 |
+
"loss": 0.3271,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5374505301216592,
|
| 160 |
+
"grad_norm": 7.60768461227417,
|
| 161 |
+
"learning_rate": 8.95692533181337e-06,
|
| 162 |
+
"loss": 0.3244,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5618800996726437,
|
| 167 |
+
"grad_norm": 7.402951717376709,
|
| 168 |
+
"learning_rate": 9.364058301441251e-06,
|
| 169 |
+
"loss": 0.301,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5863096692236283,
|
| 174 |
+
"grad_norm": 6.322145462036133,
|
| 175 |
+
"learning_rate": 9.771191271069131e-06,
|
| 176 |
+
"loss": 0.3125,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6107392387746128,
|
| 181 |
+
"grad_norm": 9.011578559875488,
|
| 182 |
+
"learning_rate": 9.988616812811544e-06,
|
| 183 |
+
"loss": 0.3154,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6351688083255973,
|
| 188 |
+
"grad_norm": 9.336881637573242,
|
| 189 |
+
"learning_rate": 9.962627800965753e-06,
|
| 190 |
+
"loss": 0.3219,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6595983778765818,
|
| 195 |
+
"grad_norm": 7.196060657501221,
|
| 196 |
+
"learning_rate": 9.936638789119961e-06,
|
| 197 |
+
"loss": 0.3062,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6840279474275663,
|
| 202 |
+
"grad_norm": 4.0668864250183105,
|
| 203 |
+
"learning_rate": 9.91064977727417e-06,
|
| 204 |
+
"loss": 0.3033,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7084575169785509,
|
| 209 |
+
"grad_norm": 8.530991554260254,
|
| 210 |
+
"learning_rate": 9.884660765428378e-06,
|
| 211 |
+
"loss": 0.2962,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7328870865295354,
|
| 216 |
+
"grad_norm": 10.066252708435059,
|
| 217 |
+
"learning_rate": 9.858671753582586e-06,
|
| 218 |
+
"loss": 0.2936,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7573166560805199,
|
| 223 |
+
"grad_norm": 7.515346050262451,
|
| 224 |
+
"learning_rate": 9.832682741736795e-06,
|
| 225 |
+
"loss": 0.2899,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7817462256315044,
|
| 230 |
+
"grad_norm": 11.316283226013184,
|
| 231 |
+
"learning_rate": 9.806693729891003e-06,
|
| 232 |
+
"loss": 0.2927,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8061757951824888,
|
| 237 |
+
"grad_norm": 14.11153793334961,
|
| 238 |
+
"learning_rate": 9.78070471804521e-06,
|
| 239 |
+
"loss": 0.2925,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8306053647334734,
|
| 244 |
+
"grad_norm": 10.594977378845215,
|
| 245 |
+
"learning_rate": 9.75471570619942e-06,
|
| 246 |
+
"loss": 0.283,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8550349342844579,
|
| 251 |
+
"grad_norm": 7.546814441680908,
|
| 252 |
+
"learning_rate": 9.728726694353629e-06,
|
| 253 |
+
"loss": 0.2922,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8794645038354424,
|
| 258 |
+
"grad_norm": 10.046064376831055,
|
| 259 |
+
"learning_rate": 9.702737682507836e-06,
|
| 260 |
+
"loss": 0.2913,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9038940733864269,
|
| 265 |
+
"grad_norm": 3.178946018218994,
|
| 266 |
+
"learning_rate": 9.676748670662046e-06,
|
| 267 |
+
"loss": 0.2824,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.9283236429374114,
|
| 272 |
+
"grad_norm": 7.698747634887695,
|
| 273 |
+
"learning_rate": 9.650759658816253e-06,
|
| 274 |
+
"loss": 0.2838,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.952753212488396,
|
| 279 |
+
"grad_norm": 5.425921440124512,
|
| 280 |
+
"learning_rate": 9.624770646970461e-06,
|
| 281 |
+
"loss": 0.2728,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.9771827820393805,
|
| 286 |
+
"grad_norm": 5.203506946563721,
|
| 287 |
+
"learning_rate": 9.59878163512467e-06,
|
| 288 |
+
"loss": 0.28,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0,
|
| 293 |
+
"eval_accuracy": 0.883413494571939,
|
| 294 |
+
"eval_averaged_scores": 0.8669552141922922,
|
| 295 |
+
"eval_f1": 0.8504969338126454,
|
| 296 |
+
"eval_loss": 0.27397170662879944,
|
| 297 |
+
"eval_runtime": 19.9293,
|
| 298 |
+
"eval_samples_per_second": 1825.705,
|
| 299 |
+
"eval_steps_per_second": 114.154,
|
| 300 |
+
"step": 20467
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.001612351590365,
|
| 304 |
+
"grad_norm": 0.704022228717804,
|
| 305 |
+
"learning_rate": 9.572792623278878e-06,
|
| 306 |
+
"loss": 0.2708,
|
| 307 |
+
"step": 20500
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.0260419211413494,
|
| 311 |
+
"grad_norm": 7.036589622497559,
|
| 312 |
+
"learning_rate": 9.546803611433086e-06,
|
| 313 |
+
"loss": 0.24,
|
| 314 |
+
"step": 21000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.050471490692334,
|
| 318 |
+
"grad_norm": 9.941107749938965,
|
| 319 |
+
"learning_rate": 9.520814599587295e-06,
|
| 320 |
+
"loss": 0.2366,
|
| 321 |
+
"step": 21500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.0749010602433184,
|
| 325 |
+
"grad_norm": 12.233573913574219,
|
| 326 |
+
"learning_rate": 9.494825587741504e-06,
|
| 327 |
+
"loss": 0.2299,
|
| 328 |
+
"step": 22000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.099330629794303,
|
| 332 |
+
"grad_norm": 5.590576171875,
|
| 333 |
+
"learning_rate": 9.468836575895712e-06,
|
| 334 |
+
"loss": 0.2342,
|
| 335 |
+
"step": 22500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.1237601993452875,
|
| 339 |
+
"grad_norm": 7.248867988586426,
|
| 340 |
+
"learning_rate": 9.442847564049921e-06,
|
| 341 |
+
"loss": 0.2393,
|
| 342 |
+
"step": 23000
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.1481897688962721,
|
| 346 |
+
"grad_norm": 7.67080545425415,
|
| 347 |
+
"learning_rate": 9.416858552204129e-06,
|
| 348 |
+
"loss": 0.2283,
|
| 349 |
+
"step": 23500
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.1726193384472565,
|
| 353 |
+
"grad_norm": 6.089188575744629,
|
| 354 |
+
"learning_rate": 9.390869540358337e-06,
|
| 355 |
+
"loss": 0.2294,
|
| 356 |
+
"step": 24000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.1970489079982412,
|
| 360 |
+
"grad_norm": 10.239951133728027,
|
| 361 |
+
"learning_rate": 9.364880528512546e-06,
|
| 362 |
+
"loss": 0.2396,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.2214784775492256,
|
| 367 |
+
"grad_norm": 4.684466361999512,
|
| 368 |
+
"learning_rate": 9.338891516666754e-06,
|
| 369 |
+
"loss": 0.2368,
|
| 370 |
+
"step": 25000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.24590804710021,
|
| 374 |
+
"grad_norm": 4.887259006500244,
|
| 375 |
+
"learning_rate": 9.312902504820961e-06,
|
| 376 |
+
"loss": 0.237,
|
| 377 |
+
"step": 25500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.2703376166511946,
|
| 381 |
+
"grad_norm": 11.124235153198242,
|
| 382 |
+
"learning_rate": 9.28691349297517e-06,
|
| 383 |
+
"loss": 0.2353,
|
| 384 |
+
"step": 26000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.2947671862021792,
|
| 388 |
+
"grad_norm": 5.592607498168945,
|
| 389 |
+
"learning_rate": 9.26092448112938e-06,
|
| 390 |
+
"loss": 0.2371,
|
| 391 |
+
"step": 26500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.3191967557531636,
|
| 395 |
+
"grad_norm": 11.883992195129395,
|
| 396 |
+
"learning_rate": 9.234935469283588e-06,
|
| 397 |
+
"loss": 0.2278,
|
| 398 |
+
"step": 27000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.343626325304148,
|
| 402 |
+
"grad_norm": 3.901329517364502,
|
| 403 |
+
"learning_rate": 9.208946457437797e-06,
|
| 404 |
+
"loss": 0.2306,
|
| 405 |
+
"step": 27500
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.3680558948551327,
|
| 409 |
+
"grad_norm": 7.284310340881348,
|
| 410 |
+
"learning_rate": 9.182957445592005e-06,
|
| 411 |
+
"loss": 0.235,
|
| 412 |
+
"step": 28000
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.392485464406117,
|
| 416 |
+
"grad_norm": 8.776345252990723,
|
| 417 |
+
"learning_rate": 9.156968433746212e-06,
|
| 418 |
+
"loss": 0.235,
|
| 419 |
+
"step": 28500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.4169150339571017,
|
| 423 |
+
"grad_norm": 8.628862380981445,
|
| 424 |
+
"learning_rate": 9.130979421900422e-06,
|
| 425 |
+
"loss": 0.2377,
|
| 426 |
+
"step": 29000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.4413446035080861,
|
| 430 |
+
"grad_norm": 6.743818759918213,
|
| 431 |
+
"learning_rate": 9.10499041005463e-06,
|
| 432 |
+
"loss": 0.237,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.4657741730590708,
|
| 437 |
+
"grad_norm": 8.978547096252441,
|
| 438 |
+
"learning_rate": 9.079001398208837e-06,
|
| 439 |
+
"loss": 0.2197,
|
| 440 |
+
"step": 30000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.4902037426100552,
|
| 444 |
+
"grad_norm": 8.523369789123535,
|
| 445 |
+
"learning_rate": 9.053012386363046e-06,
|
| 446 |
+
"loss": 0.2328,
|
| 447 |
+
"step": 30500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.5146333121610396,
|
| 451 |
+
"grad_norm": 4.043966770172119,
|
| 452 |
+
"learning_rate": 9.027023374517256e-06,
|
| 453 |
+
"loss": 0.2256,
|
| 454 |
+
"step": 31000
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 1.5390628817120242,
|
| 458 |
+
"grad_norm": 2.507680654525757,
|
| 459 |
+
"learning_rate": 9.001034362671463e-06,
|
| 460 |
+
"loss": 0.2372,
|
| 461 |
+
"step": 31500
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.5634924512630088,
|
| 465 |
+
"grad_norm": 10.360971450805664,
|
| 466 |
+
"learning_rate": 8.975045350825672e-06,
|
| 467 |
+
"loss": 0.2304,
|
| 468 |
+
"step": 32000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.5879220208139933,
|
| 472 |
+
"grad_norm": 13.895761489868164,
|
| 473 |
+
"learning_rate": 8.94905633897988e-06,
|
| 474 |
+
"loss": 0.2326,
|
| 475 |
+
"step": 32500
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.6123515903649777,
|
| 479 |
+
"grad_norm": 1.2886134386062622,
|
| 480 |
+
"learning_rate": 8.923067327134088e-06,
|
| 481 |
+
"loss": 0.2214,
|
| 482 |
+
"step": 33000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.6367811599159623,
|
| 486 |
+
"grad_norm": 5.124172210693359,
|
| 487 |
+
"learning_rate": 8.897078315288297e-06,
|
| 488 |
+
"loss": 0.2211,
|
| 489 |
+
"step": 33500
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.661210729466947,
|
| 493 |
+
"grad_norm": 13.803434371948242,
|
| 494 |
+
"learning_rate": 8.871089303442505e-06,
|
| 495 |
+
"loss": 0.2297,
|
| 496 |
+
"step": 34000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.6856402990179313,
|
| 500 |
+
"grad_norm": 8.947907447814941,
|
| 501 |
+
"learning_rate": 8.845100291596712e-06,
|
| 502 |
+
"loss": 0.2366,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.7100698685689157,
|
| 507 |
+
"grad_norm": 5.495797157287598,
|
| 508 |
+
"learning_rate": 8.819111279750922e-06,
|
| 509 |
+
"loss": 0.2256,
|
| 510 |
+
"step": 35000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.7344994381199004,
|
| 514 |
+
"grad_norm": 2.985291004180908,
|
| 515 |
+
"learning_rate": 8.793122267905131e-06,
|
| 516 |
+
"loss": 0.2268,
|
| 517 |
+
"step": 35500
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.758929007670885,
|
| 521 |
+
"grad_norm": 8.196168899536133,
|
| 522 |
+
"learning_rate": 8.767133256059339e-06,
|
| 523 |
+
"loss": 0.2299,
|
| 524 |
+
"step": 36000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.7833585772218692,
|
| 528 |
+
"grad_norm": 1.6701953411102295,
|
| 529 |
+
"learning_rate": 8.741144244213548e-06,
|
| 530 |
+
"loss": 0.22,
|
| 531 |
+
"step": 36500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 1.8077881467728538,
|
| 535 |
+
"grad_norm": 9.895888328552246,
|
| 536 |
+
"learning_rate": 8.715155232367756e-06,
|
| 537 |
+
"loss": 0.2249,
|
| 538 |
+
"step": 37000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.8322177163238385,
|
| 542 |
+
"grad_norm": 5.433578014373779,
|
| 543 |
+
"learning_rate": 8.689166220521963e-06,
|
| 544 |
+
"loss": 0.2224,
|
| 545 |
+
"step": 37500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.8566472858748229,
|
| 549 |
+
"grad_norm": 4.518404960632324,
|
| 550 |
+
"learning_rate": 8.663177208676173e-06,
|
| 551 |
+
"loss": 0.23,
|
| 552 |
+
"step": 38000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.8810768554258073,
|
| 556 |
+
"grad_norm": 18.695293426513672,
|
| 557 |
+
"learning_rate": 8.63718819683038e-06,
|
| 558 |
+
"loss": 0.2208,
|
| 559 |
+
"step": 38500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.905506424976792,
|
| 563 |
+
"grad_norm": 7.175665855407715,
|
| 564 |
+
"learning_rate": 8.61119918498459e-06,
|
| 565 |
+
"loss": 0.2241,
|
| 566 |
+
"step": 39000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.9299359945277765,
|
| 570 |
+
"grad_norm": 5.0359930992126465,
|
| 571 |
+
"learning_rate": 8.585210173138797e-06,
|
| 572 |
+
"loss": 0.2196,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.954365564078761,
|
| 577 |
+
"grad_norm": 9.02230453491211,
|
| 578 |
+
"learning_rate": 8.559221161293007e-06,
|
| 579 |
+
"loss": 0.2173,
|
| 580 |
+
"step": 40000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 1.9787951336297454,
|
| 584 |
+
"grad_norm": 17.239194869995117,
|
| 585 |
+
"learning_rate": 8.533232149447214e-06,
|
| 586 |
+
"loss": 0.2243,
|
| 587 |
+
"step": 40500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.0,
|
| 591 |
+
"eval_accuracy": 0.9014429022949018,
|
| 592 |
+
"eval_averaged_scores": 0.8850004629501047,
|
| 593 |
+
"eval_f1": 0.8685580236053075,
|
| 594 |
+
"eval_loss": 0.2654816508293152,
|
| 595 |
+
"eval_runtime": 19.8911,
|
| 596 |
+
"eval_samples_per_second": 1829.209,
|
| 597 |
+
"eval_steps_per_second": 114.373,
|
| 598 |
+
"step": 40934
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.00322470318073,
|
| 602 |
+
"grad_norm": 0.6389194130897522,
|
| 603 |
+
"learning_rate": 8.507243137601424e-06,
|
| 604 |
+
"loss": 0.2114,
|
| 605 |
+
"step": 41000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 2.0276542727317146,
|
| 609 |
+
"grad_norm": 2.8046016693115234,
|
| 610 |
+
"learning_rate": 8.481254125755631e-06,
|
| 611 |
+
"loss": 0.168,
|
| 612 |
+
"step": 41500
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.052083842282699,
|
| 616 |
+
"grad_norm": 13.7579345703125,
|
| 617 |
+
"learning_rate": 8.455265113909839e-06,
|
| 618 |
+
"loss": 0.1619,
|
| 619 |
+
"step": 42000
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 2.0765134118336834,
|
| 623 |
+
"grad_norm": 8.394590377807617,
|
| 624 |
+
"learning_rate": 8.429276102064048e-06,
|
| 625 |
+
"loss": 0.1715,
|
| 626 |
+
"step": 42500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.100942981384668,
|
| 630 |
+
"grad_norm": 12.94633674621582,
|
| 631 |
+
"learning_rate": 8.403287090218256e-06,
|
| 632 |
+
"loss": 0.1781,
|
| 633 |
+
"step": 43000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.1253725509356527,
|
| 637 |
+
"grad_norm": 15.842960357666016,
|
| 638 |
+
"learning_rate": 8.377298078372465e-06,
|
| 639 |
+
"loss": 0.178,
|
| 640 |
+
"step": 43500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.149802120486637,
|
| 644 |
+
"grad_norm": 35.52817916870117,
|
| 645 |
+
"learning_rate": 8.351309066526673e-06,
|
| 646 |
+
"loss": 0.171,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 2.1742316900376215,
|
| 651 |
+
"grad_norm": 3.224804162979126,
|
| 652 |
+
"learning_rate": 8.325320054680882e-06,
|
| 653 |
+
"loss": 0.1768,
|
| 654 |
+
"step": 44500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.198661259588606,
|
| 658 |
+
"grad_norm": 0.9365151524543762,
|
| 659 |
+
"learning_rate": 8.29933104283509e-06,
|
| 660 |
+
"loss": 0.1814,
|
| 661 |
+
"step": 45000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.2230908291395908,
|
| 665 |
+
"grad_norm": 0.5694290399551392,
|
| 666 |
+
"learning_rate": 8.2733420309893e-06,
|
| 667 |
+
"loss": 0.1732,
|
| 668 |
+
"step": 45500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.247520398690575,
|
| 672 |
+
"grad_norm": 20.7188720703125,
|
| 673 |
+
"learning_rate": 8.247353019143507e-06,
|
| 674 |
+
"loss": 0.1764,
|
| 675 |
+
"step": 46000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.2719499682415596,
|
| 679 |
+
"grad_norm": 20.063705444335938,
|
| 680 |
+
"learning_rate": 8.221364007297714e-06,
|
| 681 |
+
"loss": 0.1611,
|
| 682 |
+
"step": 46500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.2963795377925442,
|
| 686 |
+
"grad_norm": 3.708296060562134,
|
| 687 |
+
"learning_rate": 8.195374995451924e-06,
|
| 688 |
+
"loss": 0.177,
|
| 689 |
+
"step": 47000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.3208091073435284,
|
| 693 |
+
"grad_norm": 12.466748237609863,
|
| 694 |
+
"learning_rate": 8.169385983606131e-06,
|
| 695 |
+
"loss": 0.174,
|
| 696 |
+
"step": 47500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.345238676894513,
|
| 700 |
+
"grad_norm": 18.929283142089844,
|
| 701 |
+
"learning_rate": 8.14339697176034e-06,
|
| 702 |
+
"loss": 0.1738,
|
| 703 |
+
"step": 48000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.3696682464454977,
|
| 707 |
+
"grad_norm": 0.1316341757774353,
|
| 708 |
+
"learning_rate": 8.117407959914548e-06,
|
| 709 |
+
"loss": 0.1602,
|
| 710 |
+
"step": 48500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.3940978159964823,
|
| 714 |
+
"grad_norm": 0.9068466424942017,
|
| 715 |
+
"learning_rate": 8.091418948068758e-06,
|
| 716 |
+
"loss": 0.1817,
|
| 717 |
+
"step": 49000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.4185273855474665,
|
| 721 |
+
"grad_norm": 9.32274055480957,
|
| 722 |
+
"learning_rate": 8.065429936222965e-06,
|
| 723 |
+
"loss": 0.179,
|
| 724 |
+
"step": 49500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.442956955098451,
|
| 728 |
+
"grad_norm": 13.762944221496582,
|
| 729 |
+
"learning_rate": 8.039440924377175e-06,
|
| 730 |
+
"loss": 0.1767,
|
| 731 |
+
"step": 50000
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.4673865246494358,
|
| 735 |
+
"grad_norm": 15.327546119689941,
|
| 736 |
+
"learning_rate": 8.013451912531382e-06,
|
| 737 |
+
"loss": 0.179,
|
| 738 |
+
"step": 50500
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.49181609420042,
|
| 742 |
+
"grad_norm": 11.3632230758667,
|
| 743 |
+
"learning_rate": 7.98746290068559e-06,
|
| 744 |
+
"loss": 0.1871,
|
| 745 |
+
"step": 51000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.5162456637514046,
|
| 749 |
+
"grad_norm": 7.470567226409912,
|
| 750 |
+
"learning_rate": 7.9614738888398e-06,
|
| 751 |
+
"loss": 0.1788,
|
| 752 |
+
"step": 51500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.540675233302389,
|
| 756 |
+
"grad_norm": 19.140291213989258,
|
| 757 |
+
"learning_rate": 7.935484876994007e-06,
|
| 758 |
+
"loss": 0.1773,
|
| 759 |
+
"step": 52000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.565104802853374,
|
| 763 |
+
"grad_norm": 1.0424716472625732,
|
| 764 |
+
"learning_rate": 7.909495865148216e-06,
|
| 765 |
+
"loss": 0.1834,
|
| 766 |
+
"step": 52500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.5895343724043585,
|
| 770 |
+
"grad_norm": 6.240677356719971,
|
| 771 |
+
"learning_rate": 7.883506853302424e-06,
|
| 772 |
+
"loss": 0.1818,
|
| 773 |
+
"step": 53000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.6139639419553427,
|
| 777 |
+
"grad_norm": 15.357166290283203,
|
| 778 |
+
"learning_rate": 7.857517841456633e-06,
|
| 779 |
+
"loss": 0.1822,
|
| 780 |
+
"step": 53500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.6383935115063273,
|
| 784 |
+
"grad_norm": 19.889368057250977,
|
| 785 |
+
"learning_rate": 7.831528829610841e-06,
|
| 786 |
+
"loss": 0.1802,
|
| 787 |
+
"step": 54000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.662823081057312,
|
| 791 |
+
"grad_norm": 0.9387052059173584,
|
| 792 |
+
"learning_rate": 7.80553981776505e-06,
|
| 793 |
+
"loss": 0.1695,
|
| 794 |
+
"step": 54500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.687252650608296,
|
| 798 |
+
"grad_norm": 7.633642673492432,
|
| 799 |
+
"learning_rate": 7.779550805919258e-06,
|
| 800 |
+
"loss": 0.1751,
|
| 801 |
+
"step": 55000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.7116822201592807,
|
| 805 |
+
"grad_norm": 3.3056068420410156,
|
| 806 |
+
"learning_rate": 7.753561794073467e-06,
|
| 807 |
+
"loss": 0.181,
|
| 808 |
+
"step": 55500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.7361117897102654,
|
| 812 |
+
"grad_norm": 1.0076144933700562,
|
| 813 |
+
"learning_rate": 7.727572782227675e-06,
|
| 814 |
+
"loss": 0.1803,
|
| 815 |
+
"step": 56000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.76054135926125,
|
| 819 |
+
"grad_norm": 19.961585998535156,
|
| 820 |
+
"learning_rate": 7.701583770381883e-06,
|
| 821 |
+
"loss": 0.1782,
|
| 822 |
+
"step": 56500
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 2.784970928812234,
|
| 826 |
+
"grad_norm": 20.330406188964844,
|
| 827 |
+
"learning_rate": 7.675594758536092e-06,
|
| 828 |
+
"loss": 0.1806,
|
| 829 |
+
"step": 57000
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.809400498363219,
|
| 833 |
+
"grad_norm": 5.856145858764648,
|
| 834 |
+
"learning_rate": 7.6496057466903e-06,
|
| 835 |
+
"loss": 0.1861,
|
| 836 |
+
"step": 57500
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.8338300679142034,
|
| 840 |
+
"grad_norm": 8.22775650024414,
|
| 841 |
+
"learning_rate": 7.623616734844508e-06,
|
| 842 |
+
"loss": 0.1827,
|
| 843 |
+
"step": 58000
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.8582596374651876,
|
| 847 |
+
"grad_norm": 29.643142700195312,
|
| 848 |
+
"learning_rate": 7.597627722998717e-06,
|
| 849 |
+
"loss": 0.1681,
|
| 850 |
+
"step": 58500
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 2.8826892070161723,
|
| 854 |
+
"grad_norm": 0.6174699664115906,
|
| 855 |
+
"learning_rate": 7.571638711152926e-06,
|
| 856 |
+
"loss": 0.1805,
|
| 857 |
+
"step": 59000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.907118776567157,
|
| 861 |
+
"grad_norm": 3.572753429412842,
|
| 862 |
+
"learning_rate": 7.5456496993071335e-06,
|
| 863 |
+
"loss": 0.1775,
|
| 864 |
+
"step": 59500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.9315483461181415,
|
| 868 |
+
"grad_norm": 32.64729309082031,
|
| 869 |
+
"learning_rate": 7.519660687461342e-06,
|
| 870 |
+
"loss": 0.1796,
|
| 871 |
+
"step": 60000
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 2.955977915669126,
|
| 875 |
+
"grad_norm": 4.017928600311279,
|
| 876 |
+
"learning_rate": 7.4936716756155505e-06,
|
| 877 |
+
"loss": 0.1858,
|
| 878 |
+
"step": 60500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.9804074852201103,
|
| 882 |
+
"grad_norm": 38.9568977355957,
|
| 883 |
+
"learning_rate": 7.467682663769759e-06,
|
| 884 |
+
"loss": 0.1792,
|
| 885 |
+
"step": 61000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.0,
|
| 889 |
+
"eval_accuracy": 0.9041638037652879,
|
| 890 |
+
"eval_averaged_scores": 0.886470190747281,
|
| 891 |
+
"eval_f1": 0.8687765777292741,
|
| 892 |
+
"eval_loss": 0.3331463634967804,
|
| 893 |
+
"eval_runtime": 19.9629,
|
| 894 |
+
"eval_samples_per_second": 1822.629,
|
| 895 |
+
"eval_steps_per_second": 113.961,
|
| 896 |
+
"step": 61401
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 3.004837054771095,
|
| 900 |
+
"grad_norm": 26.86715316772461,
|
| 901 |
+
"learning_rate": 7.441693651923967e-06,
|
| 902 |
+
"loss": 0.1638,
|
| 903 |
+
"step": 61500
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 3.0292666243220796,
|
| 907 |
+
"grad_norm": 1.3579039573669434,
|
| 908 |
+
"learning_rate": 7.415704640078175e-06,
|
| 909 |
+
"loss": 0.1363,
|
| 910 |
+
"step": 62000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 3.053696193873064,
|
| 914 |
+
"grad_norm": 10.071706771850586,
|
| 915 |
+
"learning_rate": 7.389715628232384e-06,
|
| 916 |
+
"loss": 0.1412,
|
| 917 |
+
"step": 62500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.0781257634240484,
|
| 921 |
+
"grad_norm": 27.224559783935547,
|
| 922 |
+
"learning_rate": 7.363726616386593e-06,
|
| 923 |
+
"loss": 0.1332,
|
| 924 |
+
"step": 63000
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.102555332975033,
|
| 928 |
+
"grad_norm": 0.21357271075248718,
|
| 929 |
+
"learning_rate": 7.337737604540801e-06,
|
| 930 |
+
"loss": 0.1303,
|
| 931 |
+
"step": 63500
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 3.1269849025260177,
|
| 935 |
+
"grad_norm": 8.00867748260498,
|
| 936 |
+
"learning_rate": 7.311748592695009e-06,
|
| 937 |
+
"loss": 0.1303,
|
| 938 |
+
"step": 64000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 3.151414472077002,
|
| 942 |
+
"grad_norm": 0.21115398406982422,
|
| 943 |
+
"learning_rate": 7.2857595808492175e-06,
|
| 944 |
+
"loss": 0.1226,
|
| 945 |
+
"step": 64500
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 3.1758440416279865,
|
| 949 |
+
"grad_norm": 0.14586468040943146,
|
| 950 |
+
"learning_rate": 7.259770569003426e-06,
|
| 951 |
+
"loss": 0.1327,
|
| 952 |
+
"step": 65000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 3.200273611178971,
|
| 956 |
+
"grad_norm": 30.79446029663086,
|
| 957 |
+
"learning_rate": 7.2337815571576345e-06,
|
| 958 |
+
"loss": 0.138,
|
| 959 |
+
"step": 65500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.2247031807299553,
|
| 963 |
+
"grad_norm": 19.943998336791992,
|
| 964 |
+
"learning_rate": 7.207792545311842e-06,
|
| 965 |
+
"loss": 0.1285,
|
| 966 |
+
"step": 66000
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 3.24913275028094,
|
| 970 |
+
"grad_norm": 22.417388916015625,
|
| 971 |
+
"learning_rate": 7.181803533466051e-06,
|
| 972 |
+
"loss": 0.1471,
|
| 973 |
+
"step": 66500
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 3.2735623198319246,
|
| 977 |
+
"grad_norm": 32.429569244384766,
|
| 978 |
+
"learning_rate": 7.155814521620259e-06,
|
| 979 |
+
"loss": 0.1304,
|
| 980 |
+
"step": 67000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 3.297991889382909,
|
| 984 |
+
"grad_norm": 26.37249183654785,
|
| 985 |
+
"learning_rate": 7.1298255097744685e-06,
|
| 986 |
+
"loss": 0.1392,
|
| 987 |
+
"step": 67500
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 3.3224214589338934,
|
| 991 |
+
"grad_norm": 32.82535171508789,
|
| 992 |
+
"learning_rate": 7.103836497928677e-06,
|
| 993 |
+
"loss": 0.1274,
|
| 994 |
+
"step": 68000
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 3.346851028484878,
|
| 998 |
+
"grad_norm": 0.29850202798843384,
|
| 999 |
+
"learning_rate": 7.077847486082885e-06,
|
| 1000 |
+
"loss": 0.1456,
|
| 1001 |
+
"step": 68500
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 3.3712805980358627,
|
| 1005 |
+
"grad_norm": 14.505168914794922,
|
| 1006 |
+
"learning_rate": 7.051858474237093e-06,
|
| 1007 |
+
"loss": 0.1351,
|
| 1008 |
+
"step": 69000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 3.3957101675868473,
|
| 1012 |
+
"grad_norm": 0.11613203585147858,
|
| 1013 |
+
"learning_rate": 7.025869462391302e-06,
|
| 1014 |
+
"loss": 0.1431,
|
| 1015 |
+
"step": 69500
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 3.4201397371378315,
|
| 1019 |
+
"grad_norm": 46.42761993408203,
|
| 1020 |
+
"learning_rate": 6.99988045054551e-06,
|
| 1021 |
+
"loss": 0.1373,
|
| 1022 |
+
"step": 70000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.444569306688816,
|
| 1026 |
+
"grad_norm": 0.2621763050556183,
|
| 1027 |
+
"learning_rate": 6.973891438699718e-06,
|
| 1028 |
+
"loss": 0.1307,
|
| 1029 |
+
"step": 70500
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 3.4689988762398007,
|
| 1033 |
+
"grad_norm": 4.416637420654297,
|
| 1034 |
+
"learning_rate": 6.947902426853926e-06,
|
| 1035 |
+
"loss": 0.1452,
|
| 1036 |
+
"step": 71000
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 3.4934284457907854,
|
| 1040 |
+
"grad_norm": 11.90052604675293,
|
| 1041 |
+
"learning_rate": 6.921913415008135e-06,
|
| 1042 |
+
"loss": 0.1529,
|
| 1043 |
+
"step": 71500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.5178580153417696,
|
| 1047 |
+
"grad_norm": 8.983597755432129,
|
| 1048 |
+
"learning_rate": 6.895924403162344e-06,
|
| 1049 |
+
"loss": 0.1435,
|
| 1050 |
+
"step": 72000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 3.542287584892754,
|
| 1054 |
+
"grad_norm": 13.782587051391602,
|
| 1055 |
+
"learning_rate": 6.8699353913165525e-06,
|
| 1056 |
+
"loss": 0.149,
|
| 1057 |
+
"step": 72500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.566717154443739,
|
| 1061 |
+
"grad_norm": 38.436561584472656,
|
| 1062 |
+
"learning_rate": 6.84394637947076e-06,
|
| 1063 |
+
"loss": 0.1475,
|
| 1064 |
+
"step": 73000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.591146723994723,
|
| 1068 |
+
"grad_norm": 0.2623184025287628,
|
| 1069 |
+
"learning_rate": 6.817957367624969e-06,
|
| 1070 |
+
"loss": 0.1361,
|
| 1071 |
+
"step": 73500
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.6155762935457076,
|
| 1075 |
+
"grad_norm": 0.18694473803043365,
|
| 1076 |
+
"learning_rate": 6.791968355779177e-06,
|
| 1077 |
+
"loss": 0.1393,
|
| 1078 |
+
"step": 74000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 3.6400058630966923,
|
| 1082 |
+
"grad_norm": 0.17510893940925598,
|
| 1083 |
+
"learning_rate": 6.765979343933386e-06,
|
| 1084 |
+
"loss": 0.1447,
|
| 1085 |
+
"step": 74500
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.664435432647677,
|
| 1089 |
+
"grad_norm": 34.51966094970703,
|
| 1090 |
+
"learning_rate": 6.739990332087593e-06,
|
| 1091 |
+
"loss": 0.1332,
|
| 1092 |
+
"step": 75000
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 3.6888650021986615,
|
| 1096 |
+
"grad_norm": 38.42536544799805,
|
| 1097 |
+
"learning_rate": 6.714001320241802e-06,
|
| 1098 |
+
"loss": 0.1348,
|
| 1099 |
+
"step": 75500
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 3.7132945717496457,
|
| 1103 |
+
"grad_norm": 9.119300842285156,
|
| 1104 |
+
"learning_rate": 6.68801230839601e-06,
|
| 1105 |
+
"loss": 0.1547,
|
| 1106 |
+
"step": 76000
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.7377241413006304,
|
| 1110 |
+
"grad_norm": 5.815863132476807,
|
| 1111 |
+
"learning_rate": 6.66202329655022e-06,
|
| 1112 |
+
"loss": 0.1329,
|
| 1113 |
+
"step": 76500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 3.7621537108516145,
|
| 1117 |
+
"grad_norm": 5.0319623947143555,
|
| 1118 |
+
"learning_rate": 6.636034284704428e-06,
|
| 1119 |
+
"loss": 0.1466,
|
| 1120 |
+
"step": 77000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 3.786583280402599,
|
| 1124 |
+
"grad_norm": 16.154020309448242,
|
| 1125 |
+
"learning_rate": 6.610045272858637e-06,
|
| 1126 |
+
"loss": 0.1452,
|
| 1127 |
+
"step": 77500
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 3.811012849953584,
|
| 1131 |
+
"grad_norm": 54.372764587402344,
|
| 1132 |
+
"learning_rate": 6.584056261012844e-06,
|
| 1133 |
+
"loss": 0.1376,
|
| 1134 |
+
"step": 78000
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 3.8354424195045684,
|
| 1138 |
+
"grad_norm": 13.13427734375,
|
| 1139 |
+
"learning_rate": 6.558067249167053e-06,
|
| 1140 |
+
"loss": 0.149,
|
| 1141 |
+
"step": 78500
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 3.859871989055553,
|
| 1145 |
+
"grad_norm": 0.2712130844593048,
|
| 1146 |
+
"learning_rate": 6.532078237321261e-06,
|
| 1147 |
+
"loss": 0.1462,
|
| 1148 |
+
"step": 79000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 3.8843015586065373,
|
| 1152 |
+
"grad_norm": 0.3860025107860565,
|
| 1153 |
+
"learning_rate": 6.50608922547547e-06,
|
| 1154 |
+
"loss": 0.1459,
|
| 1155 |
+
"step": 79500
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 3.908731128157522,
|
| 1159 |
+
"grad_norm": 6.515185832977295,
|
| 1160 |
+
"learning_rate": 6.480100213629677e-06,
|
| 1161 |
+
"loss": 0.1425,
|
| 1162 |
+
"step": 80000
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 3.9331606977085065,
|
| 1166 |
+
"grad_norm": 10.842967987060547,
|
| 1167 |
+
"learning_rate": 6.454111201783886e-06,
|
| 1168 |
+
"loss": 0.1441,
|
| 1169 |
+
"step": 80500
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 3.9575902672594907,
|
| 1173 |
+
"grad_norm": 0.9704266786575317,
|
| 1174 |
+
"learning_rate": 6.428122189938095e-06,
|
| 1175 |
+
"loss": 0.1474,
|
| 1176 |
+
"step": 81000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 3.9820198368104753,
|
| 1180 |
+
"grad_norm": 7.245330333709717,
|
| 1181 |
+
"learning_rate": 6.402133178092304e-06,
|
| 1182 |
+
"loss": 0.1481,
|
| 1183 |
+
"step": 81500
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 4.0,
|
| 1187 |
+
"eval_accuracy": 0.9080939947780678,
|
| 1188 |
+
"eval_averaged_scores": 0.8914815131800832,
|
| 1189 |
+
"eval_f1": 0.8748690315820985,
|
| 1190 |
+
"eval_loss": 0.3747365474700928,
|
| 1191 |
+
"eval_runtime": 20.0359,
|
| 1192 |
+
"eval_samples_per_second": 1815.991,
|
| 1193 |
+
"eval_steps_per_second": 113.546,
|
| 1194 |
+
"step": 81868
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 4.00644940636146,
|
| 1198 |
+
"grad_norm": 0.18246012926101685,
|
| 1199 |
+
"learning_rate": 6.376144166246512e-06,
|
| 1200 |
+
"loss": 0.1331,
|
| 1201 |
+
"step": 82000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 4.030878975912445,
|
| 1205 |
+
"grad_norm": 1.72697114944458,
|
| 1206 |
+
"learning_rate": 6.35015515440072e-06,
|
| 1207 |
+
"loss": 0.1115,
|
| 1208 |
+
"step": 82500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 4.055308545463429,
|
| 1212 |
+
"grad_norm": 0.20146696269512177,
|
| 1213 |
+
"learning_rate": 6.324166142554928e-06,
|
| 1214 |
+
"loss": 0.1008,
|
| 1215 |
+
"step": 83000
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 4.079738115014414,
|
| 1219 |
+
"grad_norm": 2.5919785499572754,
|
| 1220 |
+
"learning_rate": 6.298177130709137e-06,
|
| 1221 |
+
"loss": 0.1129,
|
| 1222 |
+
"step": 83500
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 4.104167684565398,
|
| 1226 |
+
"grad_norm": 0.21622005105018616,
|
| 1227 |
+
"learning_rate": 6.272188118863345e-06,
|
| 1228 |
+
"loss": 0.111,
|
| 1229 |
+
"step": 84000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 4.128597254116382,
|
| 1233 |
+
"grad_norm": 1.2439846992492676,
|
| 1234 |
+
"learning_rate": 6.246199107017553e-06,
|
| 1235 |
+
"loss": 0.1008,
|
| 1236 |
+
"step": 84500
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 4.153026823667367,
|
| 1240 |
+
"grad_norm": 45.39404296875,
|
| 1241 |
+
"learning_rate": 6.220210095171761e-06,
|
| 1242 |
+
"loss": 0.1027,
|
| 1243 |
+
"step": 85000
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 4.1774563932183515,
|
| 1247 |
+
"grad_norm": 0.1981722116470337,
|
| 1248 |
+
"learning_rate": 6.194221083325971e-06,
|
| 1249 |
+
"loss": 0.0999,
|
| 1250 |
+
"step": 85500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 4.201885962769336,
|
| 1254 |
+
"grad_norm": 0.32059067487716675,
|
| 1255 |
+
"learning_rate": 6.168232071480179e-06,
|
| 1256 |
+
"loss": 0.1068,
|
| 1257 |
+
"step": 86000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 4.226315532320321,
|
| 1261 |
+
"grad_norm": 92.19124603271484,
|
| 1262 |
+
"learning_rate": 6.142243059634388e-06,
|
| 1263 |
+
"loss": 0.1093,
|
| 1264 |
+
"step": 86500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 4.250745101871305,
|
| 1268 |
+
"grad_norm": 3.318345069885254,
|
| 1269 |
+
"learning_rate": 6.116254047788595e-06,
|
| 1270 |
+
"loss": 0.0965,
|
| 1271 |
+
"step": 87000
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 4.275174671422289,
|
| 1275 |
+
"grad_norm": 0.1142730861902237,
|
| 1276 |
+
"learning_rate": 6.090265035942804e-06,
|
| 1277 |
+
"loss": 0.0903,
|
| 1278 |
+
"step": 87500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 4.299604240973274,
|
| 1282 |
+
"grad_norm": 40.84705352783203,
|
| 1283 |
+
"learning_rate": 6.064276024097012e-06,
|
| 1284 |
+
"loss": 0.1133,
|
| 1285 |
+
"step": 88000
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 4.324033810524258,
|
| 1289 |
+
"grad_norm": 0.1759137064218521,
|
| 1290 |
+
"learning_rate": 6.038287012251221e-06,
|
| 1291 |
+
"loss": 0.1148,
|
| 1292 |
+
"step": 88500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 4.348463380075243,
|
| 1296 |
+
"grad_norm": 0.03476906195282936,
|
| 1297 |
+
"learning_rate": 6.0122980004054285e-06,
|
| 1298 |
+
"loss": 0.1214,
|
| 1299 |
+
"step": 89000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 4.372892949626228,
|
| 1303 |
+
"grad_norm": 0.10678890347480774,
|
| 1304 |
+
"learning_rate": 5.986308988559637e-06,
|
| 1305 |
+
"loss": 0.1132,
|
| 1306 |
+
"step": 89500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 4.397322519177212,
|
| 1310 |
+
"grad_norm": 0.07550999522209167,
|
| 1311 |
+
"learning_rate": 5.9603199767138455e-06,
|
| 1312 |
+
"loss": 0.101,
|
| 1313 |
+
"step": 90000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 4.421752088728197,
|
| 1317 |
+
"grad_norm": 58.40228271484375,
|
| 1318 |
+
"learning_rate": 5.934330964868055e-06,
|
| 1319 |
+
"loss": 0.1132,
|
| 1320 |
+
"step": 90500
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 4.4461816582791815,
|
| 1324 |
+
"grad_norm": 60.99013137817383,
|
| 1325 |
+
"learning_rate": 5.908341953022263e-06,
|
| 1326 |
+
"loss": 0.1087,
|
| 1327 |
+
"step": 91000
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 4.470611227830165,
|
| 1331 |
+
"grad_norm": 8.375946044921875,
|
| 1332 |
+
"learning_rate": 5.882352941176471e-06,
|
| 1333 |
+
"loss": 0.1165,
|
| 1334 |
+
"step": 91500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 4.49504079738115,
|
| 1338 |
+
"grad_norm": 47.66468048095703,
|
| 1339 |
+
"learning_rate": 5.856363929330679e-06,
|
| 1340 |
+
"loss": 0.1082,
|
| 1341 |
+
"step": 92000
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 4.5194703669321346,
|
| 1345 |
+
"grad_norm": 53.97107696533203,
|
| 1346 |
+
"learning_rate": 5.830374917484888e-06,
|
| 1347 |
+
"loss": 0.1055,
|
| 1348 |
+
"step": 92500
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 4.543899936483119,
|
| 1352 |
+
"grad_norm": 0.5832508206367493,
|
| 1353 |
+
"learning_rate": 5.804385905639096e-06,
|
| 1354 |
+
"loss": 0.1216,
|
| 1355 |
+
"step": 93000
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 4.568329506034104,
|
| 1359 |
+
"grad_norm": 0.8978909254074097,
|
| 1360 |
+
"learning_rate": 5.778396893793304e-06,
|
| 1361 |
+
"loss": 0.1079,
|
| 1362 |
+
"step": 93500
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 4.5927590755850884,
|
| 1366 |
+
"grad_norm": 0.25415295362472534,
|
| 1367 |
+
"learning_rate": 5.7524078819475125e-06,
|
| 1368 |
+
"loss": 0.1087,
|
| 1369 |
+
"step": 94000
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 4.617188645136073,
|
| 1373 |
+
"grad_norm": 2.019859790802002,
|
| 1374 |
+
"learning_rate": 5.726418870101721e-06,
|
| 1375 |
+
"loss": 0.1127,
|
| 1376 |
+
"step": 94500
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 4.641618214687057,
|
| 1380 |
+
"grad_norm": 0.10225861519575119,
|
| 1381 |
+
"learning_rate": 5.70042985825593e-06,
|
| 1382 |
+
"loss": 0.1144,
|
| 1383 |
+
"step": 95000
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 4.6660477842380415,
|
| 1387 |
+
"grad_norm": 0.3908604383468628,
|
| 1388 |
+
"learning_rate": 5.674440846410139e-06,
|
| 1389 |
+
"loss": 0.1155,
|
| 1390 |
+
"step": 95500
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 4.690477353789026,
|
| 1394 |
+
"grad_norm": 56.87211227416992,
|
| 1395 |
+
"learning_rate": 5.6484518345643465e-06,
|
| 1396 |
+
"loss": 0.1046,
|
| 1397 |
+
"step": 96000
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 4.714906923340011,
|
| 1401 |
+
"grad_norm": 55.28709030151367,
|
| 1402 |
+
"learning_rate": 5.622462822718555e-06,
|
| 1403 |
+
"loss": 0.1182,
|
| 1404 |
+
"step": 96500
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 4.739336492890995,
|
| 1408 |
+
"grad_norm": 0.26022660732269287,
|
| 1409 |
+
"learning_rate": 5.5964738108727635e-06,
|
| 1410 |
+
"loss": 0.1132,
|
| 1411 |
+
"step": 97000
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 4.76376606244198,
|
| 1415 |
+
"grad_norm": 83.0501480102539,
|
| 1416 |
+
"learning_rate": 5.570484799026972e-06,
|
| 1417 |
+
"loss": 0.0989,
|
| 1418 |
+
"step": 97500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 4.788195631992965,
|
| 1422 |
+
"grad_norm": 1.0344722270965576,
|
| 1423 |
+
"learning_rate": 5.54449578718118e-06,
|
| 1424 |
+
"loss": 0.1154,
|
| 1425 |
+
"step": 98000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 4.812625201543948,
|
| 1429 |
+
"grad_norm": 15.780781745910645,
|
| 1430 |
+
"learning_rate": 5.518506775335388e-06,
|
| 1431 |
+
"loss": 0.1212,
|
| 1432 |
+
"step": 98500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 4.837054771094933,
|
| 1436 |
+
"grad_norm": 28.23257064819336,
|
| 1437 |
+
"learning_rate": 5.4925177634895966e-06,
|
| 1438 |
+
"loss": 0.1054,
|
| 1439 |
+
"step": 99000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 4.861484340645918,
|
| 1443 |
+
"grad_norm": 7.375139236450195,
|
| 1444 |
+
"learning_rate": 5.466528751643806e-06,
|
| 1445 |
+
"loss": 0.1206,
|
| 1446 |
+
"step": 99500
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 4.885913910196902,
|
| 1450 |
+
"grad_norm": 0.2415337711572647,
|
| 1451 |
+
"learning_rate": 5.440539739798014e-06,
|
| 1452 |
+
"loss": 0.1143,
|
| 1453 |
+
"step": 100000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 4.910343479747887,
|
| 1457 |
+
"grad_norm": 6.2803778648376465,
|
| 1458 |
+
"learning_rate": 5.414550727952223e-06,
|
| 1459 |
+
"loss": 0.1091,
|
| 1460 |
+
"step": 100500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 4.9347730492988715,
|
| 1464 |
+
"grad_norm": 0.14708127081394196,
|
| 1465 |
+
"learning_rate": 5.3885617161064305e-06,
|
| 1466 |
+
"loss": 0.1162,
|
| 1467 |
+
"step": 101000
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 4.959202618849856,
|
| 1471 |
+
"grad_norm": 118.1122817993164,
|
| 1472 |
+
"learning_rate": 5.362572704260639e-06,
|
| 1473 |
+
"loss": 0.1189,
|
| 1474 |
+
"step": 101500
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 4.98363218840084,
|
| 1478 |
+
"grad_norm": 1.3488441705703735,
|
| 1479 |
+
"learning_rate": 5.3365836924148475e-06,
|
| 1480 |
+
"loss": 0.1205,
|
| 1481 |
+
"step": 102000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 5.0,
|
| 1485 |
+
"eval_accuracy": 0.908231414044249,
|
| 1486 |
+
"eval_averaged_scores": 0.8918882665659018,
|
| 1487 |
+
"eval_f1": 0.8755451190875545,
|
| 1488 |
+
"eval_loss": 0.4310702681541443,
|
| 1489 |
+
"eval_runtime": 19.9311,
|
| 1490 |
+
"eval_samples_per_second": 1825.537,
|
| 1491 |
+
"eval_steps_per_second": 114.143,
|
| 1492 |
+
"step": 102335
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 5.0080617579518245,
|
| 1496 |
+
"grad_norm": 0.28304779529571533,
|
| 1497 |
+
"learning_rate": 5.310594680569056e-06,
|
| 1498 |
+
"loss": 0.1044,
|
| 1499 |
+
"step": 102500
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 5.032491327502809,
|
| 1503 |
+
"grad_norm": 0.2214316725730896,
|
| 1504 |
+
"learning_rate": 5.284605668723264e-06,
|
| 1505 |
+
"loss": 0.0743,
|
| 1506 |
+
"step": 103000
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 5.056920897053794,
|
| 1510 |
+
"grad_norm": 17.650388717651367,
|
| 1511 |
+
"learning_rate": 5.258616656877472e-06,
|
| 1512 |
+
"loss": 0.0765,
|
| 1513 |
+
"step": 103500
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 5.081350466604778,
|
| 1517 |
+
"grad_norm": 0.08155571669340134,
|
| 1518 |
+
"learning_rate": 5.2326276450316815e-06,
|
| 1519 |
+
"loss": 0.0766,
|
| 1520 |
+
"step": 104000
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 5.105780036155763,
|
| 1524 |
+
"grad_norm": 1.303634524345398,
|
| 1525 |
+
"learning_rate": 5.20663863318589e-06,
|
| 1526 |
+
"loss": 0.0886,
|
| 1527 |
+
"step": 104500
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 5.130209605706748,
|
| 1531 |
+
"grad_norm": 0.11499236524105072,
|
| 1532 |
+
"learning_rate": 5.1806496213400985e-06,
|
| 1533 |
+
"loss": 0.0714,
|
| 1534 |
+
"step": 105000
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 5.154639175257732,
|
| 1538 |
+
"grad_norm": 0.27438443899154663,
|
| 1539 |
+
"learning_rate": 5.154660609494306e-06,
|
| 1540 |
+
"loss": 0.0743,
|
| 1541 |
+
"step": 105500
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 5.179068744808716,
|
| 1545 |
+
"grad_norm": 5.550343990325928,
|
| 1546 |
+
"learning_rate": 5.128671597648515e-06,
|
| 1547 |
+
"loss": 0.0782,
|
| 1548 |
+
"step": 106000
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 5.203498314359701,
|
| 1552 |
+
"grad_norm": 0.1588057577610016,
|
| 1553 |
+
"learning_rate": 5.102682585802723e-06,
|
| 1554 |
+
"loss": 0.0854,
|
| 1555 |
+
"step": 106500
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 5.227927883910685,
|
| 1559 |
+
"grad_norm": 102.02461242675781,
|
| 1560 |
+
"learning_rate": 5.0766935739569316e-06,
|
| 1561 |
+
"loss": 0.0664,
|
| 1562 |
+
"step": 107000
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 5.25235745346167,
|
| 1566 |
+
"grad_norm": 10.157758712768555,
|
| 1567 |
+
"learning_rate": 5.050704562111139e-06,
|
| 1568 |
+
"loss": 0.0971,
|
| 1569 |
+
"step": 107500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 5.276787023012655,
|
| 1573 |
+
"grad_norm": 0.3199716806411743,
|
| 1574 |
+
"learning_rate": 5.024715550265348e-06,
|
| 1575 |
+
"loss": 0.0845,
|
| 1576 |
+
"step": 108000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 5.301216592563639,
|
| 1580 |
+
"grad_norm": 0.11773143708705902,
|
| 1581 |
+
"learning_rate": 4.998726538419556e-06,
|
| 1582 |
+
"loss": 0.0799,
|
| 1583 |
+
"step": 108500
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 5.325646162114624,
|
| 1587 |
+
"grad_norm": 0.045085687190294266,
|
| 1588 |
+
"learning_rate": 4.972737526573765e-06,
|
| 1589 |
+
"loss": 0.0831,
|
| 1590 |
+
"step": 109000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 5.3500757316656085,
|
| 1594 |
+
"grad_norm": 0.12273421883583069,
|
| 1595 |
+
"learning_rate": 4.946748514727974e-06,
|
| 1596 |
+
"loss": 0.0809,
|
| 1597 |
+
"step": 109500
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 5.374505301216592,
|
| 1601 |
+
"grad_norm": 0.06519081443548203,
|
| 1602 |
+
"learning_rate": 4.920759502882182e-06,
|
| 1603 |
+
"loss": 0.0724,
|
| 1604 |
+
"step": 110000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 5.398934870767577,
|
| 1608 |
+
"grad_norm": 0.09597588330507278,
|
| 1609 |
+
"learning_rate": 4.89477049103639e-06,
|
| 1610 |
+
"loss": 0.0853,
|
| 1611 |
+
"step": 110500
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 5.4233644403185615,
|
| 1615 |
+
"grad_norm": 0.03921746462583542,
|
| 1616 |
+
"learning_rate": 4.868781479190599e-06,
|
| 1617 |
+
"loss": 0.0733,
|
| 1618 |
+
"step": 111000
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 5.447794009869546,
|
| 1622 |
+
"grad_norm": 10.62073040008545,
|
| 1623 |
+
"learning_rate": 4.842792467344807e-06,
|
| 1624 |
+
"loss": 0.0945,
|
| 1625 |
+
"step": 111500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 5.472223579420531,
|
| 1629 |
+
"grad_norm": 0.6295870542526245,
|
| 1630 |
+
"learning_rate": 4.816803455499016e-06,
|
| 1631 |
+
"loss": 0.0832,
|
| 1632 |
+
"step": 112000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 5.496653148971515,
|
| 1636 |
+
"grad_norm": 0.025005605071783066,
|
| 1637 |
+
"learning_rate": 4.790814443653224e-06,
|
| 1638 |
+
"loss": 0.0794,
|
| 1639 |
+
"step": 112500
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 5.5210827185225,
|
| 1643 |
+
"grad_norm": 0.026888804510235786,
|
| 1644 |
+
"learning_rate": 4.764825431807433e-06,
|
| 1645 |
+
"loss": 0.0863,
|
| 1646 |
+
"step": 113000
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 5.545512288073484,
|
| 1650 |
+
"grad_norm": 0.2212212085723877,
|
| 1651 |
+
"learning_rate": 4.73883641996164e-06,
|
| 1652 |
+
"loss": 0.0844,
|
| 1653 |
+
"step": 113500
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 5.569941857624468,
|
| 1657 |
+
"grad_norm": 0.07905617356300354,
|
| 1658 |
+
"learning_rate": 4.712847408115849e-06,
|
| 1659 |
+
"loss": 0.0857,
|
| 1660 |
+
"step": 114000
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 5.594371427175453,
|
| 1664 |
+
"grad_norm": 0.3286093473434448,
|
| 1665 |
+
"learning_rate": 4.686858396270057e-06,
|
| 1666 |
+
"loss": 0.0924,
|
| 1667 |
+
"step": 114500
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 5.618800996726438,
|
| 1671 |
+
"grad_norm": 0.14107511937618256,
|
| 1672 |
+
"learning_rate": 4.660869384424266e-06,
|
| 1673 |
+
"loss": 0.0775,
|
| 1674 |
+
"step": 115000
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 5.643230566277422,
|
| 1678 |
+
"grad_norm": 9.034845352172852,
|
| 1679 |
+
"learning_rate": 4.634880372578474e-06,
|
| 1680 |
+
"loss": 0.0763,
|
| 1681 |
+
"step": 115500
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 5.667660135828407,
|
| 1685 |
+
"grad_norm": 45.53312301635742,
|
| 1686 |
+
"learning_rate": 4.608891360732683e-06,
|
| 1687 |
+
"loss": 0.0833,
|
| 1688 |
+
"step": 116000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 5.6920897053793915,
|
| 1692 |
+
"grad_norm": 0.6340117454528809,
|
| 1693 |
+
"learning_rate": 4.582902348886891e-06,
|
| 1694 |
+
"loss": 0.0847,
|
| 1695 |
+
"step": 116500
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 5.716519274930375,
|
| 1699 |
+
"grad_norm": 3.8106794357299805,
|
| 1700 |
+
"learning_rate": 4.5569133370411e-06,
|
| 1701 |
+
"loss": 0.0856,
|
| 1702 |
+
"step": 117000
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 5.74094884448136,
|
| 1706 |
+
"grad_norm": 0.541683554649353,
|
| 1707 |
+
"learning_rate": 4.530924325195308e-06,
|
| 1708 |
+
"loss": 0.0762,
|
| 1709 |
+
"step": 117500
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 5.7653784140323445,
|
| 1713 |
+
"grad_norm": 0.7401725649833679,
|
| 1714 |
+
"learning_rate": 4.504935313349516e-06,
|
| 1715 |
+
"loss": 0.0787,
|
| 1716 |
+
"step": 118000
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 5.789807983583329,
|
| 1720 |
+
"grad_norm": 4.942069053649902,
|
| 1721 |
+
"learning_rate": 4.478946301503724e-06,
|
| 1722 |
+
"loss": 0.0879,
|
| 1723 |
+
"step": 118500
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 5.814237553134314,
|
| 1727 |
+
"grad_norm": 0.025865860283374786,
|
| 1728 |
+
"learning_rate": 4.452957289657933e-06,
|
| 1729 |
+
"loss": 0.0767,
|
| 1730 |
+
"step": 119000
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 5.838667122685298,
|
| 1734 |
+
"grad_norm": 0.19948020577430725,
|
| 1735 |
+
"learning_rate": 4.426968277812141e-06,
|
| 1736 |
+
"loss": 0.0764,
|
| 1737 |
+
"step": 119500
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 5.863096692236283,
|
| 1741 |
+
"grad_norm": 0.23808695375919342,
|
| 1742 |
+
"learning_rate": 4.40097926596635e-06,
|
| 1743 |
+
"loss": 0.0768,
|
| 1744 |
+
"step": 120000
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 5.887526261787268,
|
| 1748 |
+
"grad_norm": 0.1378837376832962,
|
| 1749 |
+
"learning_rate": 4.374990254120558e-06,
|
| 1750 |
+
"loss": 0.0873,
|
| 1751 |
+
"step": 120500
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 5.911955831338252,
|
| 1755 |
+
"grad_norm": 0.013579963706433773,
|
| 1756 |
+
"learning_rate": 4.349001242274767e-06,
|
| 1757 |
+
"loss": 0.0767,
|
| 1758 |
+
"step": 121000
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 5.936385400889236,
|
| 1762 |
+
"grad_norm": 0.11858664453029633,
|
| 1763 |
+
"learning_rate": 4.323012230428975e-06,
|
| 1764 |
+
"loss": 0.091,
|
| 1765 |
+
"step": 121500
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 5.960814970440221,
|
| 1769 |
+
"grad_norm": 13.749309539794922,
|
| 1770 |
+
"learning_rate": 4.297023218583184e-06,
|
| 1771 |
+
"loss": 0.0845,
|
| 1772 |
+
"step": 122000
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 5.985244539991205,
|
| 1776 |
+
"grad_norm": 0.19270148873329163,
|
| 1777 |
+
"learning_rate": 4.271034206737391e-06,
|
| 1778 |
+
"loss": 0.0866,
|
| 1779 |
+
"step": 122500
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 6.0,
|
| 1783 |
+
"eval_accuracy": 0.9059777380788787,
|
| 1784 |
+
"eval_averaged_scores": 0.8906733019878164,
|
| 1785 |
+
"eval_f1": 0.875368865896754,
|
| 1786 |
+
"eval_loss": 0.500464677810669,
|
| 1787 |
+
"eval_runtime": 19.9132,
|
| 1788 |
+
"eval_samples_per_second": 1827.184,
|
| 1789 |
+
"eval_steps_per_second": 114.246,
|
| 1790 |
+
"step": 122802
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 6.00967410954219,
|
| 1794 |
+
"grad_norm": 0.0461135059595108,
|
| 1795 |
+
"learning_rate": 4.2450451948916e-06,
|
| 1796 |
+
"loss": 0.0779,
|
| 1797 |
+
"step": 123000
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 6.034103679093175,
|
| 1801 |
+
"grad_norm": 0.028277171775698662,
|
| 1802 |
+
"learning_rate": 4.219056183045809e-06,
|
| 1803 |
+
"loss": 0.0554,
|
| 1804 |
+
"step": 123500
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 6.058533248644159,
|
| 1808 |
+
"grad_norm": 0.015514826402068138,
|
| 1809 |
+
"learning_rate": 4.193067171200017e-06,
|
| 1810 |
+
"loss": 0.0584,
|
| 1811 |
+
"step": 124000
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 6.082962818195144,
|
| 1815 |
+
"grad_norm": 0.13774634897708893,
|
| 1816 |
+
"learning_rate": 4.167078159354225e-06,
|
| 1817 |
+
"loss": 0.0616,
|
| 1818 |
+
"step": 124500
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 6.107392387746128,
|
| 1822 |
+
"grad_norm": 0.047606222331523895,
|
| 1823 |
+
"learning_rate": 4.141089147508434e-06,
|
| 1824 |
+
"loss": 0.0541,
|
| 1825 |
+
"step": 125000
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 6.131821957297112,
|
| 1829 |
+
"grad_norm": 0.03313114866614342,
|
| 1830 |
+
"learning_rate": 4.115100135662642e-06,
|
| 1831 |
+
"loss": 0.0606,
|
| 1832 |
+
"step": 125500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 6.156251526848097,
|
| 1836 |
+
"grad_norm": 0.08608361333608627,
|
| 1837 |
+
"learning_rate": 4.089111123816851e-06,
|
| 1838 |
+
"loss": 0.0592,
|
| 1839 |
+
"step": 126000
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 6.1806810963990815,
|
| 1843 |
+
"grad_norm": 76.78373718261719,
|
| 1844 |
+
"learning_rate": 4.063122111971059e-06,
|
| 1845 |
+
"loss": 0.0478,
|
| 1846 |
+
"step": 126500
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 6.205110665950066,
|
| 1850 |
+
"grad_norm": 0.04603453725576401,
|
| 1851 |
+
"learning_rate": 4.037133100125267e-06,
|
| 1852 |
+
"loss": 0.0588,
|
| 1853 |
+
"step": 127000
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 6.229540235501051,
|
| 1857 |
+
"grad_norm": 0.018753621727228165,
|
| 1858 |
+
"learning_rate": 4.011144088279475e-06,
|
| 1859 |
+
"loss": 0.0518,
|
| 1860 |
+
"step": 127500
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 6.253969805052035,
|
| 1864 |
+
"grad_norm": 0.013871916569769382,
|
| 1865 |
+
"learning_rate": 3.985155076433685e-06,
|
| 1866 |
+
"loss": 0.0604,
|
| 1867 |
+
"step": 128000
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 6.278399374603019,
|
| 1871 |
+
"grad_norm": 0.023428503423929214,
|
| 1872 |
+
"learning_rate": 3.959166064587892e-06,
|
| 1873 |
+
"loss": 0.059,
|
| 1874 |
+
"step": 128500
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 6.302828944154004,
|
| 1878 |
+
"grad_norm": 8.137282371520996,
|
| 1879 |
+
"learning_rate": 3.933177052742101e-06,
|
| 1880 |
+
"loss": 0.0647,
|
| 1881 |
+
"step": 129000
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 6.327258513704988,
|
| 1885 |
+
"grad_norm": 0.03562696650624275,
|
| 1886 |
+
"learning_rate": 3.907188040896309e-06,
|
| 1887 |
+
"loss": 0.0577,
|
| 1888 |
+
"step": 129500
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 6.351688083255973,
|
| 1892 |
+
"grad_norm": 0.0467020608484745,
|
| 1893 |
+
"learning_rate": 3.881199029050518e-06,
|
| 1894 |
+
"loss": 0.063,
|
| 1895 |
+
"step": 130000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 6.376117652806958,
|
| 1899 |
+
"grad_norm": 0.03905618563294411,
|
| 1900 |
+
"learning_rate": 3.855210017204726e-06,
|
| 1901 |
+
"loss": 0.0612,
|
| 1902 |
+
"step": 130500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 6.400547222357942,
|
| 1906 |
+
"grad_norm": 0.027326637879014015,
|
| 1907 |
+
"learning_rate": 3.829221005358935e-06,
|
| 1908 |
+
"loss": 0.0594,
|
| 1909 |
+
"step": 131000
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 6.424976791908927,
|
| 1913 |
+
"grad_norm": 0.010337826795876026,
|
| 1914 |
+
"learning_rate": 3.803231993513143e-06,
|
| 1915 |
+
"loss": 0.0627,
|
| 1916 |
+
"step": 131500
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 6.449406361459911,
|
| 1920 |
+
"grad_norm": 0.026293253526091576,
|
| 1921 |
+
"learning_rate": 3.777242981667351e-06,
|
| 1922 |
+
"loss": 0.0553,
|
| 1923 |
+
"step": 132000
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 6.473835931010895,
|
| 1927 |
+
"grad_norm": 39.016990661621094,
|
| 1928 |
+
"learning_rate": 3.75125396982156e-06,
|
| 1929 |
+
"loss": 0.0715,
|
| 1930 |
+
"step": 132500
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 6.49826550056188,
|
| 1934 |
+
"grad_norm": 0.01873265765607357,
|
| 1935 |
+
"learning_rate": 3.7252649579757684e-06,
|
| 1936 |
+
"loss": 0.0581,
|
| 1937 |
+
"step": 133000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 6.5226950701128645,
|
| 1941 |
+
"grad_norm": 0.03351232409477234,
|
| 1942 |
+
"learning_rate": 3.6992759461299764e-06,
|
| 1943 |
+
"loss": 0.0648,
|
| 1944 |
+
"step": 133500
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 6.547124639663849,
|
| 1948 |
+
"grad_norm": 0.020647812634706497,
|
| 1949 |
+
"learning_rate": 3.673286934284185e-06,
|
| 1950 |
+
"loss": 0.0571,
|
| 1951 |
+
"step": 134000
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 6.571554209214834,
|
| 1955 |
+
"grad_norm": 0.004467953462153673,
|
| 1956 |
+
"learning_rate": 3.647297922438393e-06,
|
| 1957 |
+
"loss": 0.0569,
|
| 1958 |
+
"step": 134500
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 6.595983778765818,
|
| 1962 |
+
"grad_norm": 0.03921693563461304,
|
| 1963 |
+
"learning_rate": 3.621308910592602e-06,
|
| 1964 |
+
"loss": 0.0657,
|
| 1965 |
+
"step": 135000
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 6.620413348316802,
|
| 1969 |
+
"grad_norm": 12.03000545501709,
|
| 1970 |
+
"learning_rate": 3.5953198987468104e-06,
|
| 1971 |
+
"loss": 0.0657,
|
| 1972 |
+
"step": 135500
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 6.644842917867787,
|
| 1976 |
+
"grad_norm": 0.023439181968569756,
|
| 1977 |
+
"learning_rate": 3.5693308869010185e-06,
|
| 1978 |
+
"loss": 0.0735,
|
| 1979 |
+
"step": 136000
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 6.669272487418771,
|
| 1983 |
+
"grad_norm": 3.436170816421509,
|
| 1984 |
+
"learning_rate": 3.543341875055227e-06,
|
| 1985 |
+
"loss": 0.0656,
|
| 1986 |
+
"step": 136500
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 6.693702056969756,
|
| 1990 |
+
"grad_norm": 0.009104551747441292,
|
| 1991 |
+
"learning_rate": 3.5173528632094355e-06,
|
| 1992 |
+
"loss": 0.0655,
|
| 1993 |
+
"step": 137000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 6.718131626520741,
|
| 1997 |
+
"grad_norm": 173.34230041503906,
|
| 1998 |
+
"learning_rate": 3.491363851363644e-06,
|
| 1999 |
+
"loss": 0.0621,
|
| 2000 |
+
"step": 137500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 6.742561196071725,
|
| 2004 |
+
"grad_norm": 134.16038513183594,
|
| 2005 |
+
"learning_rate": 3.465374839517852e-06,
|
| 2006 |
+
"loss": 0.0676,
|
| 2007 |
+
"step": 138000
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 6.76699076562271,
|
| 2011 |
+
"grad_norm": 61.84733200073242,
|
| 2012 |
+
"learning_rate": 3.4393858276720605e-06,
|
| 2013 |
+
"loss": 0.061,
|
| 2014 |
+
"step": 138500
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 6.791420335173695,
|
| 2018 |
+
"grad_norm": 0.04754955321550369,
|
| 2019 |
+
"learning_rate": 3.4133968158262686e-06,
|
| 2020 |
+
"loss": 0.0664,
|
| 2021 |
+
"step": 139000
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 6.815849904724679,
|
| 2025 |
+
"grad_norm": 0.05394621938467026,
|
| 2026 |
+
"learning_rate": 3.3874078039804775e-06,
|
| 2027 |
+
"loss": 0.0644,
|
| 2028 |
+
"step": 139500
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 6.840279474275663,
|
| 2032 |
+
"grad_norm": 0.04520469903945923,
|
| 2033 |
+
"learning_rate": 3.361418792134686e-06,
|
| 2034 |
+
"loss": 0.0667,
|
| 2035 |
+
"step": 140000
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 6.864709043826648,
|
| 2039 |
+
"grad_norm": 15.504779815673828,
|
| 2040 |
+
"learning_rate": 3.335429780288894e-06,
|
| 2041 |
+
"loss": 0.06,
|
| 2042 |
+
"step": 140500
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 6.889138613377632,
|
| 2046 |
+
"grad_norm": 169.57215881347656,
|
| 2047 |
+
"learning_rate": 3.3094407684431025e-06,
|
| 2048 |
+
"loss": 0.0578,
|
| 2049 |
+
"step": 141000
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 6.913568182928617,
|
| 2053 |
+
"grad_norm": 0.2180391252040863,
|
| 2054 |
+
"learning_rate": 3.283451756597311e-06,
|
| 2055 |
+
"loss": 0.0539,
|
| 2056 |
+
"step": 141500
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 6.9379977524796015,
|
| 2060 |
+
"grad_norm": 0.11775655299425125,
|
| 2061 |
+
"learning_rate": 3.2574627447515195e-06,
|
| 2062 |
+
"loss": 0.0663,
|
| 2063 |
+
"step": 142000
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 6.962427322030586,
|
| 2067 |
+
"grad_norm": 0.1945144534111023,
|
| 2068 |
+
"learning_rate": 3.2314737329057276e-06,
|
| 2069 |
+
"loss": 0.0668,
|
| 2070 |
+
"step": 142500
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 6.986856891581571,
|
| 2074 |
+
"grad_norm": 0.1719103306531906,
|
| 2075 |
+
"learning_rate": 3.205484721059936e-06,
|
| 2076 |
+
"loss": 0.0606,
|
| 2077 |
+
"step": 143000
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 7.0,
|
| 2081 |
+
"eval_accuracy": 0.9077916723924694,
|
| 2082 |
+
"eval_averaged_scores": 0.8922162391818581,
|
| 2083 |
+
"eval_f1": 0.8766408059712468,
|
| 2084 |
+
"eval_loss": 0.5274677872657776,
|
| 2085 |
+
"eval_runtime": 20.1172,
|
| 2086 |
+
"eval_samples_per_second": 1808.655,
|
| 2087 |
+
"eval_steps_per_second": 113.088,
|
| 2088 |
+
"step": 143269
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 7.0112864611325545,
|
| 2092 |
+
"grad_norm": 44.23752975463867,
|
| 2093 |
+
"learning_rate": 3.179495709214144e-06,
|
| 2094 |
+
"loss": 0.061,
|
| 2095 |
+
"step": 143500
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 7.035716030683539,
|
| 2099 |
+
"grad_norm": 7.647191524505615,
|
| 2100 |
+
"learning_rate": 3.153506697368353e-06,
|
| 2101 |
+
"loss": 0.0452,
|
| 2102 |
+
"step": 144000
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 7.060145600234524,
|
| 2106 |
+
"grad_norm": 0.0981924906373024,
|
| 2107 |
+
"learning_rate": 3.1275176855225615e-06,
|
| 2108 |
+
"loss": 0.0485,
|
| 2109 |
+
"step": 144500
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 7.084575169785508,
|
| 2113 |
+
"grad_norm": 0.0033761723898351192,
|
| 2114 |
+
"learning_rate": 3.1015286736767696e-06,
|
| 2115 |
+
"loss": 0.037,
|
| 2116 |
+
"step": 145000
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 7.109004739336493,
|
| 2120 |
+
"grad_norm": 0.08983515948057175,
|
| 2121 |
+
"learning_rate": 3.075539661830978e-06,
|
| 2122 |
+
"loss": 0.0362,
|
| 2123 |
+
"step": 145500
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 7.133434308887478,
|
| 2127 |
+
"grad_norm": 0.03911528363823891,
|
| 2128 |
+
"learning_rate": 3.049550649985186e-06,
|
| 2129 |
+
"loss": 0.0366,
|
| 2130 |
+
"step": 146000
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 7.157863878438462,
|
| 2134 |
+
"grad_norm": 0.07024548947811127,
|
| 2135 |
+
"learning_rate": 3.023561638139395e-06,
|
| 2136 |
+
"loss": 0.0464,
|
| 2137 |
+
"step": 146500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 7.182293447989446,
|
| 2141 |
+
"grad_norm": 0.030780920758843422,
|
| 2142 |
+
"learning_rate": 2.9975726262936036e-06,
|
| 2143 |
+
"loss": 0.0339,
|
| 2144 |
+
"step": 147000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 7.206723017540431,
|
| 2148 |
+
"grad_norm": 6.837735652923584,
|
| 2149 |
+
"learning_rate": 2.9715836144478116e-06,
|
| 2150 |
+
"loss": 0.0456,
|
| 2151 |
+
"step": 147500
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 7.231152587091415,
|
| 2155 |
+
"grad_norm": 1.052836537361145,
|
| 2156 |
+
"learning_rate": 2.94559460260202e-06,
|
| 2157 |
+
"loss": 0.0538,
|
| 2158 |
+
"step": 148000
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 7.2555821566424,
|
| 2162 |
+
"grad_norm": 0.13163608312606812,
|
| 2163 |
+
"learning_rate": 2.9196055907562286e-06,
|
| 2164 |
+
"loss": 0.053,
|
| 2165 |
+
"step": 148500
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 7.2800117261933845,
|
| 2169 |
+
"grad_norm": 3.5442745685577393,
|
| 2170 |
+
"learning_rate": 2.893616578910437e-06,
|
| 2171 |
+
"loss": 0.0451,
|
| 2172 |
+
"step": 149000
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 7.304441295744369,
|
| 2176 |
+
"grad_norm": 0.02228480763733387,
|
| 2177 |
+
"learning_rate": 2.867627567064645e-06,
|
| 2178 |
+
"loss": 0.0489,
|
| 2179 |
+
"step": 149500
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 7.328870865295354,
|
| 2183 |
+
"grad_norm": 0.06422539055347443,
|
| 2184 |
+
"learning_rate": 2.8416385552188536e-06,
|
| 2185 |
+
"loss": 0.0479,
|
| 2186 |
+
"step": 150000
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 7.3533004348463376,
|
| 2190 |
+
"grad_norm": 0.008064317516982555,
|
| 2191 |
+
"learning_rate": 2.8156495433730617e-06,
|
| 2192 |
+
"loss": 0.0387,
|
| 2193 |
+
"step": 150500
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 7.377730004397322,
|
| 2197 |
+
"grad_norm": 0.1824285238981247,
|
| 2198 |
+
"learning_rate": 2.7896605315272706e-06,
|
| 2199 |
+
"loss": 0.0522,
|
| 2200 |
+
"step": 151000
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 7.402159573948307,
|
| 2204 |
+
"grad_norm": 0.06451237201690674,
|
| 2205 |
+
"learning_rate": 2.763671519681479e-06,
|
| 2206 |
+
"loss": 0.0419,
|
| 2207 |
+
"step": 151500
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 7.4265891434992914,
|
| 2211 |
+
"grad_norm": 0.032122429460287094,
|
| 2212 |
+
"learning_rate": 2.737682507835687e-06,
|
| 2213 |
+
"loss": 0.0427,
|
| 2214 |
+
"step": 152000
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 7.451018713050276,
|
| 2218 |
+
"grad_norm": 0.006546043790876865,
|
| 2219 |
+
"learning_rate": 2.7116934959898957e-06,
|
| 2220 |
+
"loss": 0.0453,
|
| 2221 |
+
"step": 152500
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 7.475448282601261,
|
| 2225 |
+
"grad_norm": 59.247711181640625,
|
| 2226 |
+
"learning_rate": 2.685704484144104e-06,
|
| 2227 |
+
"loss": 0.0533,
|
| 2228 |
+
"step": 153000
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 7.499877852152245,
|
| 2232 |
+
"grad_norm": 0.0045631807297468185,
|
| 2233 |
+
"learning_rate": 2.6597154722983127e-06,
|
| 2234 |
+
"loss": 0.0452,
|
| 2235 |
+
"step": 153500
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 7.52430742170323,
|
| 2239 |
+
"grad_norm": 0.14977307617664337,
|
| 2240 |
+
"learning_rate": 2.6337264604525207e-06,
|
| 2241 |
+
"loss": 0.0523,
|
| 2242 |
+
"step": 154000
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 7.548736991254214,
|
| 2246 |
+
"grad_norm": 0.09788861125707626,
|
| 2247 |
+
"learning_rate": 2.607737448606729e-06,
|
| 2248 |
+
"loss": 0.0426,
|
| 2249 |
+
"step": 154500
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 7.573166560805198,
|
| 2253 |
+
"grad_norm": 163.57232666015625,
|
| 2254 |
+
"learning_rate": 2.5817484367609373e-06,
|
| 2255 |
+
"loss": 0.0424,
|
| 2256 |
+
"step": 155000
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 7.597596130356183,
|
| 2260 |
+
"grad_norm": 0.013917704112827778,
|
| 2261 |
+
"learning_rate": 2.555759424915146e-06,
|
| 2262 |
+
"loss": 0.0464,
|
| 2263 |
+
"step": 155500
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 7.622025699907168,
|
| 2267 |
+
"grad_norm": 0.06623460352420807,
|
| 2268 |
+
"learning_rate": 2.5297704130693547e-06,
|
| 2269 |
+
"loss": 0.0374,
|
| 2270 |
+
"step": 156000
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 7.646455269458152,
|
| 2274 |
+
"grad_norm": 60.03496551513672,
|
| 2275 |
+
"learning_rate": 2.5037814012235627e-06,
|
| 2276 |
+
"loss": 0.0452,
|
| 2277 |
+
"step": 156500
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 7.670884839009137,
|
| 2281 |
+
"grad_norm": 0.24344667792320251,
|
| 2282 |
+
"learning_rate": 2.4777923893777712e-06,
|
| 2283 |
+
"loss": 0.0454,
|
| 2284 |
+
"step": 157000
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 7.6953144085601215,
|
| 2288 |
+
"grad_norm": 20.285869598388672,
|
| 2289 |
+
"learning_rate": 2.4518033775319797e-06,
|
| 2290 |
+
"loss": 0.0399,
|
| 2291 |
+
"step": 157500
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 7.719743978111106,
|
| 2295 |
+
"grad_norm": 0.1362639218568802,
|
| 2296 |
+
"learning_rate": 2.425814365686188e-06,
|
| 2297 |
+
"loss": 0.0367,
|
| 2298 |
+
"step": 158000
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 7.74417354766209,
|
| 2302 |
+
"grad_norm": 2.487563371658325,
|
| 2303 |
+
"learning_rate": 2.3998253538403967e-06,
|
| 2304 |
+
"loss": 0.0439,
|
| 2305 |
+
"step": 158500
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 7.7686031172130745,
|
| 2309 |
+
"grad_norm": 269.9028625488281,
|
| 2310 |
+
"learning_rate": 2.3738363419946048e-06,
|
| 2311 |
+
"loss": 0.0388,
|
| 2312 |
+
"step": 159000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 7.793032686764059,
|
| 2316 |
+
"grad_norm": 9.254499435424805,
|
| 2317 |
+
"learning_rate": 2.3478473301488133e-06,
|
| 2318 |
+
"loss": 0.0488,
|
| 2319 |
+
"step": 159500
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 7.817462256315044,
|
| 2323 |
+
"grad_norm": 0.007625956553965807,
|
| 2324 |
+
"learning_rate": 2.3218583183030217e-06,
|
| 2325 |
+
"loss": 0.0473,
|
| 2326 |
+
"step": 160000
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 7.841891825866028,
|
| 2330 |
+
"grad_norm": 4.9155402183532715,
|
| 2331 |
+
"learning_rate": 2.2958693064572302e-06,
|
| 2332 |
+
"loss": 0.0499,
|
| 2333 |
+
"step": 160500
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 7.866321395417013,
|
| 2337 |
+
"grad_norm": 25.9799861907959,
|
| 2338 |
+
"learning_rate": 2.2698802946114383e-06,
|
| 2339 |
+
"loss": 0.0483,
|
| 2340 |
+
"step": 161000
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 7.890750964967998,
|
| 2344 |
+
"grad_norm": 0.026556404307484627,
|
| 2345 |
+
"learning_rate": 2.243891282765647e-06,
|
| 2346 |
+
"loss": 0.0374,
|
| 2347 |
+
"step": 161500
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 7.915180534518981,
|
| 2351 |
+
"grad_norm": 0.009880514815449715,
|
| 2352 |
+
"learning_rate": 2.2179022709198553e-06,
|
| 2353 |
+
"loss": 0.0439,
|
| 2354 |
+
"step": 162000
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 7.939610104069966,
|
| 2358 |
+
"grad_norm": 0.005332610569894314,
|
| 2359 |
+
"learning_rate": 2.1919132590740634e-06,
|
| 2360 |
+
"loss": 0.0381,
|
| 2361 |
+
"step": 162500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 7.964039673620951,
|
| 2365 |
+
"grad_norm": 0.022613270208239555,
|
| 2366 |
+
"learning_rate": 2.1659242472282723e-06,
|
| 2367 |
+
"loss": 0.0448,
|
| 2368 |
+
"step": 163000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 7.988469243171935,
|
| 2372 |
+
"grad_norm": 0.0017537542153149843,
|
| 2373 |
+
"learning_rate": 2.1399352353824803e-06,
|
| 2374 |
+
"loss": 0.0423,
|
| 2375 |
+
"step": 163500
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 8.0,
|
| 2379 |
+
"eval_accuracy": 0.908726123402501,
|
| 2380 |
+
"eval_averaged_scores": 0.8931284118506033,
|
| 2381 |
+
"eval_f1": 0.8775307002987056,
|
| 2382 |
+
"eval_loss": 0.6607664823532104,
|
| 2383 |
+
"eval_runtime": 19.9709,
|
| 2384 |
+
"eval_samples_per_second": 1821.9,
|
| 2385 |
+
"eval_steps_per_second": 113.916,
|
| 2386 |
+
"step": 163736
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 8.01289881272292,
|
| 2390 |
+
"grad_norm": 0.002813127124682069,
|
| 2391 |
+
"learning_rate": 2.113946223536689e-06,
|
| 2392 |
+
"loss": 0.0374,
|
| 2393 |
+
"step": 164000
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 8.037328382273904,
|
| 2397 |
+
"grad_norm": 0.07825087010860443,
|
| 2398 |
+
"learning_rate": 2.0879572116908973e-06,
|
| 2399 |
+
"loss": 0.0275,
|
| 2400 |
+
"step": 164500
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 8.06175795182489,
|
| 2404 |
+
"grad_norm": 0.7584686279296875,
|
| 2405 |
+
"learning_rate": 2.061968199845106e-06,
|
| 2406 |
+
"loss": 0.0426,
|
| 2407 |
+
"step": 165000
|
| 2408 |
+
},
|
| 2409 |
+
{
|
| 2410 |
+
"epoch": 8.086187521375873,
|
| 2411 |
+
"grad_norm": 68.23661041259766,
|
| 2412 |
+
"learning_rate": 2.035979187999314e-06,
|
| 2413 |
+
"loss": 0.0291,
|
| 2414 |
+
"step": 165500
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 8.110617090926858,
|
| 2418 |
+
"grad_norm": 1.332459568977356,
|
| 2419 |
+
"learning_rate": 2.0099901761535224e-06,
|
| 2420 |
+
"loss": 0.0337,
|
| 2421 |
+
"step": 166000
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"epoch": 8.135046660477842,
|
| 2425 |
+
"grad_norm": 0.011477336287498474,
|
| 2426 |
+
"learning_rate": 1.984001164307731e-06,
|
| 2427 |
+
"loss": 0.0252,
|
| 2428 |
+
"step": 166500
|
| 2429 |
+
},
|
| 2430 |
+
{
|
| 2431 |
+
"epoch": 8.159476230028828,
|
| 2432 |
+
"grad_norm": 0.011217381805181503,
|
| 2433 |
+
"learning_rate": 1.9580121524619393e-06,
|
| 2434 |
+
"loss": 0.0291,
|
| 2435 |
+
"step": 167000
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 8.183905799579811,
|
| 2439 |
+
"grad_norm": 0.06326530873775482,
|
| 2440 |
+
"learning_rate": 1.932023140616148e-06,
|
| 2441 |
+
"loss": 0.0422,
|
| 2442 |
+
"step": 167500
|
| 2443 |
+
},
|
| 2444 |
+
{
|
| 2445 |
+
"epoch": 8.208335369130795,
|
| 2446 |
+
"grad_norm": 0.04990832507610321,
|
| 2447 |
+
"learning_rate": 1.9060341287703559e-06,
|
| 2448 |
+
"loss": 0.0292,
|
| 2449 |
+
"step": 168000
|
| 2450 |
+
},
|
| 2451 |
+
{
|
| 2452 |
+
"epoch": 8.23276493868178,
|
| 2453 |
+
"grad_norm": 0.5803171992301941,
|
| 2454 |
+
"learning_rate": 1.8800451169245646e-06,
|
| 2455 |
+
"loss": 0.0273,
|
| 2456 |
+
"step": 168500
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 8.257194508232764,
|
| 2460 |
+
"grad_norm": 27.482540130615234,
|
| 2461 |
+
"learning_rate": 1.8540561050787729e-06,
|
| 2462 |
+
"loss": 0.0413,
|
| 2463 |
+
"step": 169000
|
| 2464 |
+
},
|
| 2465 |
+
{
|
| 2466 |
+
"epoch": 8.28162407778375,
|
| 2467 |
+
"grad_norm": 0.09522279351949692,
|
| 2468 |
+
"learning_rate": 1.8280670932329814e-06,
|
| 2469 |
+
"loss": 0.0329,
|
| 2470 |
+
"step": 169500
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 8.306053647334734,
|
| 2474 |
+
"grad_norm": 24.168445587158203,
|
| 2475 |
+
"learning_rate": 1.8020780813871896e-06,
|
| 2476 |
+
"loss": 0.0269,
|
| 2477 |
+
"step": 170000
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 8.33048321688572,
|
| 2481 |
+
"grad_norm": 0.0035793723072856665,
|
| 2482 |
+
"learning_rate": 1.776089069541398e-06,
|
| 2483 |
+
"loss": 0.0356,
|
| 2484 |
+
"step": 170500
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 8.354912786436703,
|
| 2488 |
+
"grad_norm": 0.05237244442105293,
|
| 2489 |
+
"learning_rate": 1.7501000576956064e-06,
|
| 2490 |
+
"loss": 0.0357,
|
| 2491 |
+
"step": 171000
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 8.379342355987687,
|
| 2495 |
+
"grad_norm": 0.006013878155499697,
|
| 2496 |
+
"learning_rate": 1.7241110458498147e-06,
|
| 2497 |
+
"loss": 0.0334,
|
| 2498 |
+
"step": 171500
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 8.403771925538672,
|
| 2502 |
+
"grad_norm": 0.05959504470229149,
|
| 2503 |
+
"learning_rate": 1.6981220340040234e-06,
|
| 2504 |
+
"loss": 0.0311,
|
| 2505 |
+
"step": 172000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 8.428201495089656,
|
| 2509 |
+
"grad_norm": 0.009634653106331825,
|
| 2510 |
+
"learning_rate": 1.6721330221582317e-06,
|
| 2511 |
+
"loss": 0.0269,
|
| 2512 |
+
"step": 172500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 8.452631064640642,
|
| 2516 |
+
"grad_norm": 0.01209759246557951,
|
| 2517 |
+
"learning_rate": 1.6461440103124402e-06,
|
| 2518 |
+
"loss": 0.0269,
|
| 2519 |
+
"step": 173000
|
| 2520 |
+
},
|
| 2521 |
+
{
|
| 2522 |
+
"epoch": 8.477060634191625,
|
| 2523 |
+
"grad_norm": 0.005159119609743357,
|
| 2524 |
+
"learning_rate": 1.6201549984666484e-06,
|
| 2525 |
+
"loss": 0.0378,
|
| 2526 |
+
"step": 173500
|
| 2527 |
+
},
|
| 2528 |
+
{
|
| 2529 |
+
"epoch": 8.50149020374261,
|
| 2530 |
+
"grad_norm": 0.08292187750339508,
|
| 2531 |
+
"learning_rate": 1.5941659866208567e-06,
|
| 2532 |
+
"loss": 0.0336,
|
| 2533 |
+
"step": 174000
|
| 2534 |
+
},
|
| 2535 |
+
{
|
| 2536 |
+
"epoch": 8.525919773293595,
|
| 2537 |
+
"grad_norm": 22.53067970275879,
|
| 2538 |
+
"learning_rate": 1.5681769747750652e-06,
|
| 2539 |
+
"loss": 0.0357,
|
| 2540 |
+
"step": 174500
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 8.550349342844578,
|
| 2544 |
+
"grad_norm": 17.8717041015625,
|
| 2545 |
+
"learning_rate": 1.5421879629292735e-06,
|
| 2546 |
+
"loss": 0.0341,
|
| 2547 |
+
"step": 175000
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 8.574778912395564,
|
| 2551 |
+
"grad_norm": 0.6152280569076538,
|
| 2552 |
+
"learning_rate": 1.5161989510834822e-06,
|
| 2553 |
+
"loss": 0.039,
|
| 2554 |
+
"step": 175500
|
| 2555 |
+
},
|
| 2556 |
+
{
|
| 2557 |
+
"epoch": 8.599208481946548,
|
| 2558 |
+
"grad_norm": 0.0008640658925287426,
|
| 2559 |
+
"learning_rate": 1.4902099392376905e-06,
|
| 2560 |
+
"loss": 0.0415,
|
| 2561 |
+
"step": 176000
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 8.623638051497533,
|
| 2565 |
+
"grad_norm": 0.46269872784614563,
|
| 2566 |
+
"learning_rate": 1.464220927391899e-06,
|
| 2567 |
+
"loss": 0.0323,
|
| 2568 |
+
"step": 176500
|
| 2569 |
+
},
|
| 2570 |
+
{
|
| 2571 |
+
"epoch": 8.648067621048517,
|
| 2572 |
+
"grad_norm": 0.18344636261463165,
|
| 2573 |
+
"learning_rate": 1.4382319155461072e-06,
|
| 2574 |
+
"loss": 0.0317,
|
| 2575 |
+
"step": 177000
|
| 2576 |
+
},
|
| 2577 |
+
{
|
| 2578 |
+
"epoch": 8.672497190599502,
|
| 2579 |
+
"grad_norm": 0.0006130241672508419,
|
| 2580 |
+
"learning_rate": 1.4122429037003157e-06,
|
| 2581 |
+
"loss": 0.034,
|
| 2582 |
+
"step": 177500
|
| 2583 |
+
},
|
| 2584 |
+
{
|
| 2585 |
+
"epoch": 8.696926760150486,
|
| 2586 |
+
"grad_norm": 0.05424381420016289,
|
| 2587 |
+
"learning_rate": 1.386253891854524e-06,
|
| 2588 |
+
"loss": 0.0377,
|
| 2589 |
+
"step": 178000
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 8.721356329701472,
|
| 2593 |
+
"grad_norm": 0.04327189177274704,
|
| 2594 |
+
"learning_rate": 1.3602648800087323e-06,
|
| 2595 |
+
"loss": 0.0358,
|
| 2596 |
+
"step": 178500
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 8.745785899252455,
|
| 2600 |
+
"grad_norm": 0.019710328429937363,
|
| 2601 |
+
"learning_rate": 1.3342758681629408e-06,
|
| 2602 |
+
"loss": 0.0395,
|
| 2603 |
+
"step": 179000
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 8.770215468803439,
|
| 2607 |
+
"grad_norm": 1.1634199619293213,
|
| 2608 |
+
"learning_rate": 1.308286856317149e-06,
|
| 2609 |
+
"loss": 0.0308,
|
| 2610 |
+
"step": 179500
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 8.794645038354425,
|
| 2614 |
+
"grad_norm": 0.0033998938743025064,
|
| 2615 |
+
"learning_rate": 1.2822978444713577e-06,
|
| 2616 |
+
"loss": 0.0317,
|
| 2617 |
+
"step": 180000
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 8.819074607905408,
|
| 2621 |
+
"grad_norm": 0.04000772163271904,
|
| 2622 |
+
"learning_rate": 1.256308832625566e-06,
|
| 2623 |
+
"loss": 0.0317,
|
| 2624 |
+
"step": 180500
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 8.843504177456394,
|
| 2628 |
+
"grad_norm": 0.19507430493831635,
|
| 2629 |
+
"learning_rate": 1.2303198207797743e-06,
|
| 2630 |
+
"loss": 0.0393,
|
| 2631 |
+
"step": 181000
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"epoch": 8.867933747007378,
|
| 2635 |
+
"grad_norm": 0.18581752479076385,
|
| 2636 |
+
"learning_rate": 1.2043308089339828e-06,
|
| 2637 |
+
"loss": 0.0229,
|
| 2638 |
+
"step": 181500
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 8.892363316558363,
|
| 2642 |
+
"grad_norm": 15.190007209777832,
|
| 2643 |
+
"learning_rate": 1.1783417970881913e-06,
|
| 2644 |
+
"loss": 0.0283,
|
| 2645 |
+
"step": 182000
|
| 2646 |
+
},
|
| 2647 |
+
{
|
| 2648 |
+
"epoch": 8.916792886109347,
|
| 2649 |
+
"grad_norm": 0.005882304161787033,
|
| 2650 |
+
"learning_rate": 1.1523527852423996e-06,
|
| 2651 |
+
"loss": 0.0355,
|
| 2652 |
+
"step": 182500
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 8.94122245566033,
|
| 2656 |
+
"grad_norm": 0.0683102160692215,
|
| 2657 |
+
"learning_rate": 1.126363773396608e-06,
|
| 2658 |
+
"loss": 0.0433,
|
| 2659 |
+
"step": 183000
|
| 2660 |
+
},
|
| 2661 |
+
{
|
| 2662 |
+
"epoch": 8.965652025211316,
|
| 2663 |
+
"grad_norm": 0.026383813470602036,
|
| 2664 |
+
"learning_rate": 1.1003747615508165e-06,
|
| 2665 |
+
"loss": 0.0332,
|
| 2666 |
+
"step": 183500
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 8.9900815947623,
|
| 2670 |
+
"grad_norm": 0.4436836540699005,
|
| 2671 |
+
"learning_rate": 1.0743857497050248e-06,
|
| 2672 |
+
"loss": 0.0385,
|
| 2673 |
+
"step": 184000
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 9.0,
|
| 2677 |
+
"eval_accuracy": 0.9086986395492648,
|
| 2678 |
+
"eval_averaged_scores": 0.893008009485837,
|
| 2679 |
+
"eval_f1": 0.8773173794224093,
|
| 2680 |
+
"eval_loss": 0.6683973073959351,
|
| 2681 |
+
"eval_runtime": 19.8532,
|
| 2682 |
+
"eval_samples_per_second": 1832.702,
|
| 2683 |
+
"eval_steps_per_second": 114.591,
|
| 2684 |
+
"step": 184203
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 9.014511164313285,
|
| 2688 |
+
"grad_norm": 0.025027230381965637,
|
| 2689 |
+
"learning_rate": 1.048396737859233e-06,
|
| 2690 |
+
"loss": 0.0252,
|
| 2691 |
+
"step": 184500
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 9.038940733864269,
|
| 2695 |
+
"grad_norm": 278.2423400878906,
|
| 2696 |
+
"learning_rate": 1.0224077260134416e-06,
|
| 2697 |
+
"loss": 0.0261,
|
| 2698 |
+
"step": 185000
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 9.063370303415255,
|
| 2702 |
+
"grad_norm": 0.023968037217855453,
|
| 2703 |
+
"learning_rate": 9.9641871416765e-07,
|
| 2704 |
+
"loss": 0.0213,
|
| 2705 |
+
"step": 185500
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 9.087799872966238,
|
| 2709 |
+
"grad_norm": 0.0030007651075720787,
|
| 2710 |
+
"learning_rate": 9.704297023218584e-07,
|
| 2711 |
+
"loss": 0.0222,
|
| 2712 |
+
"step": 186000
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 9.112229442517222,
|
| 2716 |
+
"grad_norm": 0.08472041040658951,
|
| 2717 |
+
"learning_rate": 9.444406904760668e-07,
|
| 2718 |
+
"loss": 0.0277,
|
| 2719 |
+
"step": 186500
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 9.136659012068208,
|
| 2723 |
+
"grad_norm": 0.018034188076853752,
|
| 2724 |
+
"learning_rate": 9.184516786302752e-07,
|
| 2725 |
+
"loss": 0.0266,
|
| 2726 |
+
"step": 187000
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 9.161088581619191,
|
| 2730 |
+
"grad_norm": 0.0007292185327969491,
|
| 2731 |
+
"learning_rate": 8.924626667844836e-07,
|
| 2732 |
+
"loss": 0.0191,
|
| 2733 |
+
"step": 187500
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 9.185518151170177,
|
| 2737 |
+
"grad_norm": 0.0054621621966362,
|
| 2738 |
+
"learning_rate": 8.664736549386919e-07,
|
| 2739 |
+
"loss": 0.0244,
|
| 2740 |
+
"step": 188000
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 9.20994772072116,
|
| 2744 |
+
"grad_norm": 0.030511168763041496,
|
| 2745 |
+
"learning_rate": 8.404846430929004e-07,
|
| 2746 |
+
"loss": 0.0251,
|
| 2747 |
+
"step": 188500
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 9.234377290272146,
|
| 2751 |
+
"grad_norm": 0.0641358271241188,
|
| 2752 |
+
"learning_rate": 8.144956312471088e-07,
|
| 2753 |
+
"loss": 0.0281,
|
| 2754 |
+
"step": 189000
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 9.25880685982313,
|
| 2758 |
+
"grad_norm": 55.69062042236328,
|
| 2759 |
+
"learning_rate": 7.885066194013171e-07,
|
| 2760 |
+
"loss": 0.0208,
|
| 2761 |
+
"step": 189500
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 9.283236429374114,
|
| 2765 |
+
"grad_norm": 0.014215814881026745,
|
| 2766 |
+
"learning_rate": 7.625176075555256e-07,
|
| 2767 |
+
"loss": 0.0284,
|
| 2768 |
+
"step": 190000
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 9.3076659989251,
|
| 2772 |
+
"grad_norm": 0.005864777602255344,
|
| 2773 |
+
"learning_rate": 7.36528595709734e-07,
|
| 2774 |
+
"loss": 0.0217,
|
| 2775 |
+
"step": 190500
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 9.332095568476083,
|
| 2779 |
+
"grad_norm": 0.00025711164926178753,
|
| 2780 |
+
"learning_rate": 7.105395838639424e-07,
|
| 2781 |
+
"loss": 0.021,
|
| 2782 |
+
"step": 191000
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 9.356525138027068,
|
| 2786 |
+
"grad_norm": 0.013914588838815689,
|
| 2787 |
+
"learning_rate": 6.845505720181508e-07,
|
| 2788 |
+
"loss": 0.0234,
|
| 2789 |
+
"step": 191500
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 9.380954707578052,
|
| 2793 |
+
"grad_norm": 16.358308792114258,
|
| 2794 |
+
"learning_rate": 6.585615601723591e-07,
|
| 2795 |
+
"loss": 0.0259,
|
| 2796 |
+
"step": 192000
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 9.405384277129038,
|
| 2800 |
+
"grad_norm": 0.003568566171452403,
|
| 2801 |
+
"learning_rate": 6.325725483265676e-07,
|
| 2802 |
+
"loss": 0.0319,
|
| 2803 |
+
"step": 192500
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 9.429813846680021,
|
| 2807 |
+
"grad_norm": 0.002073039533570409,
|
| 2808 |
+
"learning_rate": 6.065835364807759e-07,
|
| 2809 |
+
"loss": 0.0196,
|
| 2810 |
+
"step": 193000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 9.454243416231005,
|
| 2814 |
+
"grad_norm": 0.09922459721565247,
|
| 2815 |
+
"learning_rate": 5.805945246349843e-07,
|
| 2816 |
+
"loss": 0.0263,
|
| 2817 |
+
"step": 193500
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 9.47867298578199,
|
| 2821 |
+
"grad_norm": 0.0005966913304291666,
|
| 2822 |
+
"learning_rate": 5.546055127891928e-07,
|
| 2823 |
+
"loss": 0.0178,
|
| 2824 |
+
"step": 194000
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 9.503102555332974,
|
| 2828 |
+
"grad_norm": 0.000937746255658567,
|
| 2829 |
+
"learning_rate": 5.286165009434012e-07,
|
| 2830 |
+
"loss": 0.0202,
|
| 2831 |
+
"step": 194500
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 9.52753212488396,
|
| 2835 |
+
"grad_norm": 0.07347114384174347,
|
| 2836 |
+
"learning_rate": 5.026274890976096e-07,
|
| 2837 |
+
"loss": 0.0206,
|
| 2838 |
+
"step": 195000
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 9.551961694434944,
|
| 2842 |
+
"grad_norm": 0.0010983546962961555,
|
| 2843 |
+
"learning_rate": 4.7663847725181796e-07,
|
| 2844 |
+
"loss": 0.0205,
|
| 2845 |
+
"step": 195500
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 9.57639126398593,
|
| 2849 |
+
"grad_norm": 0.004537857137620449,
|
| 2850 |
+
"learning_rate": 4.5064946540602635e-07,
|
| 2851 |
+
"loss": 0.0243,
|
| 2852 |
+
"step": 196000
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 9.600820833536913,
|
| 2856 |
+
"grad_norm": 0.005785027518868446,
|
| 2857 |
+
"learning_rate": 4.246604535602348e-07,
|
| 2858 |
+
"loss": 0.0221,
|
| 2859 |
+
"step": 196500
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 9.625250403087897,
|
| 2863 |
+
"grad_norm": 0.002915257355198264,
|
| 2864 |
+
"learning_rate": 3.986714417144431e-07,
|
| 2865 |
+
"loss": 0.0236,
|
| 2866 |
+
"step": 197000
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 9.649679972638882,
|
| 2870 |
+
"grad_norm": 0.006192653439939022,
|
| 2871 |
+
"learning_rate": 3.7268242986865155e-07,
|
| 2872 |
+
"loss": 0.0204,
|
| 2873 |
+
"step": 197500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 9.674109542189866,
|
| 2877 |
+
"grad_norm": 0.0025680630933493376,
|
| 2878 |
+
"learning_rate": 3.4669341802286e-07,
|
| 2879 |
+
"loss": 0.0218,
|
| 2880 |
+
"step": 198000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 9.698539111740851,
|
| 2884 |
+
"grad_norm": 0.0032936306670308113,
|
| 2885 |
+
"learning_rate": 3.207044061770684e-07,
|
| 2886 |
+
"loss": 0.0281,
|
| 2887 |
+
"step": 198500
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 9.722968681291835,
|
| 2891 |
+
"grad_norm": 0.015488212928175926,
|
| 2892 |
+
"learning_rate": 2.9471539433127676e-07,
|
| 2893 |
+
"loss": 0.0274,
|
| 2894 |
+
"step": 199000
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 9.74739825084282,
|
| 2898 |
+
"grad_norm": 0.000830151024274528,
|
| 2899 |
+
"learning_rate": 2.6872638248548514e-07,
|
| 2900 |
+
"loss": 0.0222,
|
| 2901 |
+
"step": 199500
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 9.771827820393804,
|
| 2905 |
+
"grad_norm": 0.005330337677150965,
|
| 2906 |
+
"learning_rate": 2.427373706396936e-07,
|
| 2907 |
+
"loss": 0.0214,
|
| 2908 |
+
"step": 200000
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 9.796257389944788,
|
| 2912 |
+
"grad_norm": 0.009580313228070736,
|
| 2913 |
+
"learning_rate": 2.1674835879390194e-07,
|
| 2914 |
+
"loss": 0.0233,
|
| 2915 |
+
"step": 200500
|
| 2916 |
+
},
|
| 2917 |
+
{
|
| 2918 |
+
"epoch": 9.820686959495774,
|
| 2919 |
+
"grad_norm": 0.003088373225182295,
|
| 2920 |
+
"learning_rate": 1.9075934694811035e-07,
|
| 2921 |
+
"loss": 0.0278,
|
| 2922 |
+
"step": 201000
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 9.845116529046757,
|
| 2926 |
+
"grad_norm": 0.005164031870663166,
|
| 2927 |
+
"learning_rate": 1.6477033510231873e-07,
|
| 2928 |
+
"loss": 0.0248,
|
| 2929 |
+
"step": 201500
|
| 2930 |
+
},
|
| 2931 |
+
{
|
| 2932 |
+
"epoch": 9.869546098597743,
|
| 2933 |
+
"grad_norm": 0.06639529019594193,
|
| 2934 |
+
"learning_rate": 1.3878132325652717e-07,
|
| 2935 |
+
"loss": 0.0204,
|
| 2936 |
+
"step": 202000
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 9.893975668148727,
|
| 2940 |
+
"grad_norm": 0.018493203446269035,
|
| 2941 |
+
"learning_rate": 1.1279231141073555e-07,
|
| 2942 |
+
"loss": 0.0227,
|
| 2943 |
+
"step": 202500
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 9.918405237699712,
|
| 2947 |
+
"grad_norm": 0.06634573638439178,
|
| 2948 |
+
"learning_rate": 8.680329956494395e-08,
|
| 2949 |
+
"loss": 0.0178,
|
| 2950 |
+
"step": 203000
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 9.942834807250696,
|
| 2954 |
+
"grad_norm": 0.15537451207637787,
|
| 2955 |
+
"learning_rate": 6.081428771915235e-08,
|
| 2956 |
+
"loss": 0.0189,
|
| 2957 |
+
"step": 203500
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 9.967264376801682,
|
| 2961 |
+
"grad_norm": 0.007867376320064068,
|
| 2962 |
+
"learning_rate": 3.4825275873360744e-08,
|
| 2963 |
+
"loss": 0.031,
|
| 2964 |
+
"step": 204000
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 9.991693946352665,
|
| 2968 |
+
"grad_norm": 0.0018433391815051436,
|
| 2969 |
+
"learning_rate": 8.836264027569144e-09,
|
| 2970 |
+
"loss": 0.0194,
|
| 2971 |
+
"step": 204500
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 10.0,
|
| 2975 |
+
"eval_accuracy": 0.9081489624845404,
|
| 2976 |
+
"eval_averaged_scores": 0.8924639805417318,
|
| 2977 |
+
"eval_f1": 0.8767789985989234,
|
| 2978 |
+
"eval_loss": 0.7485349774360657,
|
| 2979 |
+
"eval_runtime": 19.8865,
|
| 2980 |
+
"eval_samples_per_second": 1829.631,
|
| 2981 |
+
"eval_steps_per_second": 114.399,
|
| 2982 |
+
"step": 204670
|
| 2983 |
+
},
|
| 2984 |
+
{
|
| 2985 |
+
"epoch": 10.0,
|
| 2986 |
+
"step": 204670,
|
| 2987 |
+
"total_flos": 1.0069269492791874e+17,
|
| 2988 |
+
"train_loss": 0.1257473332849194,
|
| 2989 |
+
"train_runtime": 9921.5767,
|
| 2990 |
+
"train_samples_per_second": 330.049,
|
| 2991 |
+
"train_steps_per_second": 20.629
|
| 2992 |
+
}
|
| 2993 |
+
],
|
| 2994 |
+
"logging_steps": 500,
|
| 2995 |
+
"max_steps": 204670,
|
| 2996 |
+
"num_input_tokens_seen": 0,
|
| 2997 |
+
"num_train_epochs": 10,
|
| 2998 |
+
"save_steps": 500,
|
| 2999 |
+
"stateful_callbacks": {
|
| 3000 |
+
"TrainerControl": {
|
| 3001 |
+
"args": {
|
| 3002 |
+
"should_epoch_stop": false,
|
| 3003 |
+
"should_evaluate": false,
|
| 3004 |
+
"should_log": false,
|
| 3005 |
+
"should_save": true,
|
| 3006 |
+
"should_training_stop": true
|
| 3007 |
+
},
|
| 3008 |
+
"attributes": {}
|
| 3009 |
+
}
|
| 3010 |
+
},
|
| 3011 |
+
"total_flos": 1.0069269492791874e+17,
|
| 3012 |
+
"train_batch_size": 16,
|
| 3013 |
+
"trial_name": null,
|
| 3014 |
+
"trial_params": null
|
| 3015 |
+
}
|
qqp/bert-base-uncased_lr1e-05/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1e35c83b67b72b7aa30b8d745f9f4a74338e31c351c497cdebf27e0cb3c9deb
|
| 3 |
+
size 5240
|
qqp/bert-base-uncased_lr1e-05/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/roberta-base_lr1e-05/classifier_head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69107b1d995c1f8dfa32a0de5583a684ba5eea9405049ef539296ecd6e2591c0
|
| 3 |
+
size 2371920
|
qqp/roberta-base_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/shikexuan/nlu_model/roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.45.2",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
qqp/roberta-base_lr1e-05/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/roberta-base_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d32c08a87330f2b99f4129ac57e3add1b6de06fff4ab32382034c1332ad9c73e
|
| 3 |
+
size 498612824
|
qqp/roberta-base_lr1e-05/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
qqp/roberta-base_lr1e-05/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/roberta-base_lr1e-05/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"mask_token": "<mask>",
|
| 51 |
+
"model_max_length": 512,
|
| 52 |
+
"pad_token": "<pad>",
|
| 53 |
+
"sep_token": "</s>",
|
| 54 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 55 |
+
"trim_offsets": true,
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
qqp/roberta-base_lr1e-05/trainer_state.json
ADDED
|
@@ -0,0 +1,3015 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8973993679198728,
|
| 3 |
+
"best_model_checkpoint": "./nlu_finetuned_models/qqp/roberta-base_lr1e-05/checkpoint-204670",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 204670,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.024429569550984513,
|
| 13 |
+
"grad_norm": 10.54019546508789,
|
| 14 |
+
"learning_rate": 4.0713296962788053e-07,
|
| 15 |
+
"loss": 0.6758,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.048859139101969026,
|
| 20 |
+
"grad_norm": 7.6315178871154785,
|
| 21 |
+
"learning_rate": 8.142659392557611e-07,
|
| 22 |
+
"loss": 0.6406,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07328870865295353,
|
| 27 |
+
"grad_norm": 25.296419143676758,
|
| 28 |
+
"learning_rate": 1.2213989088836414e-06,
|
| 29 |
+
"loss": 0.528,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.09771827820393805,
|
| 34 |
+
"grad_norm": 134.22874450683594,
|
| 35 |
+
"learning_rate": 1.6285318785115221e-06,
|
| 36 |
+
"loss": 0.4561,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.12214784775492256,
|
| 41 |
+
"grad_norm": 34.93831253051758,
|
| 42 |
+
"learning_rate": 2.0356648481394024e-06,
|
| 43 |
+
"loss": 0.4151,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14657741730590707,
|
| 48 |
+
"grad_norm": 15.966097831726074,
|
| 49 |
+
"learning_rate": 2.442797817767283e-06,
|
| 50 |
+
"loss": 0.3976,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.17100698685689159,
|
| 55 |
+
"grad_norm": 19.514278411865234,
|
| 56 |
+
"learning_rate": 2.8499307873951637e-06,
|
| 57 |
+
"loss": 0.377,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1954365564078761,
|
| 62 |
+
"grad_norm": 15.483463287353516,
|
| 63 |
+
"learning_rate": 3.2570637570230442e-06,
|
| 64 |
+
"loss": 0.3688,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2198661259588606,
|
| 69 |
+
"grad_norm": 17.194320678710938,
|
| 70 |
+
"learning_rate": 3.6641967266509243e-06,
|
| 71 |
+
"loss": 0.373,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.24429569550984512,
|
| 76 |
+
"grad_norm": 7.317613124847412,
|
| 77 |
+
"learning_rate": 4.071329696278805e-06,
|
| 78 |
+
"loss": 0.3518,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2687252650608296,
|
| 83 |
+
"grad_norm": 15.082125663757324,
|
| 84 |
+
"learning_rate": 4.478462665906685e-06,
|
| 85 |
+
"loss": 0.3587,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.29315483461181413,
|
| 90 |
+
"grad_norm": 49.410526275634766,
|
| 91 |
+
"learning_rate": 4.885595635534566e-06,
|
| 92 |
+
"loss": 0.3513,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.31758440416279865,
|
| 97 |
+
"grad_norm": 52.64762878417969,
|
| 98 |
+
"learning_rate": 5.292728605162446e-06,
|
| 99 |
+
"loss": 0.3426,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.34201397371378317,
|
| 104 |
+
"grad_norm": 21.950075149536133,
|
| 105 |
+
"learning_rate": 5.6998615747903275e-06,
|
| 106 |
+
"loss": 0.3401,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3664435432647677,
|
| 111 |
+
"grad_norm": 14.155437469482422,
|
| 112 |
+
"learning_rate": 6.106994544418208e-06,
|
| 113 |
+
"loss": 0.3343,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3908731128157522,
|
| 118 |
+
"grad_norm": 12.090170860290527,
|
| 119 |
+
"learning_rate": 6.5141275140460884e-06,
|
| 120 |
+
"loss": 0.3163,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4153026823667367,
|
| 125 |
+
"grad_norm": 17.292564392089844,
|
| 126 |
+
"learning_rate": 6.921260483673968e-06,
|
| 127 |
+
"loss": 0.33,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4397322519177212,
|
| 132 |
+
"grad_norm": 29.74077033996582,
|
| 133 |
+
"learning_rate": 7.3283934533018485e-06,
|
| 134 |
+
"loss": 0.3326,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4641618214687057,
|
| 139 |
+
"grad_norm": 16.858999252319336,
|
| 140 |
+
"learning_rate": 7.73552642292973e-06,
|
| 141 |
+
"loss": 0.3229,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.48859139101969024,
|
| 146 |
+
"grad_norm": 16.417327880859375,
|
| 147 |
+
"learning_rate": 8.14265939255761e-06,
|
| 148 |
+
"loss": 0.3207,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5130209605706747,
|
| 153 |
+
"grad_norm": 18.313976287841797,
|
| 154 |
+
"learning_rate": 8.54979236218549e-06,
|
| 155 |
+
"loss": 0.3154,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5374505301216592,
|
| 160 |
+
"grad_norm": 26.479129791259766,
|
| 161 |
+
"learning_rate": 8.95692533181337e-06,
|
| 162 |
+
"loss": 0.3199,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5618800996726437,
|
| 167 |
+
"grad_norm": 11.478958129882812,
|
| 168 |
+
"learning_rate": 9.364058301441251e-06,
|
| 169 |
+
"loss": 0.3041,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5863096692236283,
|
| 174 |
+
"grad_norm": 18.56060028076172,
|
| 175 |
+
"learning_rate": 9.771191271069131e-06,
|
| 176 |
+
"loss": 0.3027,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6107392387746128,
|
| 181 |
+
"grad_norm": 7.155969142913818,
|
| 182 |
+
"learning_rate": 9.988616812811544e-06,
|
| 183 |
+
"loss": 0.3122,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6351688083255973,
|
| 188 |
+
"grad_norm": 8.636488914489746,
|
| 189 |
+
"learning_rate": 9.962627800965753e-06,
|
| 190 |
+
"loss": 0.3218,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6595983778765818,
|
| 195 |
+
"grad_norm": 23.951885223388672,
|
| 196 |
+
"learning_rate": 9.936638789119961e-06,
|
| 197 |
+
"loss": 0.3109,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6840279474275663,
|
| 202 |
+
"grad_norm": 12.705269813537598,
|
| 203 |
+
"learning_rate": 9.91064977727417e-06,
|
| 204 |
+
"loss": 0.3054,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7084575169785509,
|
| 209 |
+
"grad_norm": 19.542213439941406,
|
| 210 |
+
"learning_rate": 9.884660765428378e-06,
|
| 211 |
+
"loss": 0.2983,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7328870865295354,
|
| 216 |
+
"grad_norm": 18.7448673248291,
|
| 217 |
+
"learning_rate": 9.858671753582586e-06,
|
| 218 |
+
"loss": 0.2882,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7573166560805199,
|
| 223 |
+
"grad_norm": 12.782718658447266,
|
| 224 |
+
"learning_rate": 9.832682741736795e-06,
|
| 225 |
+
"loss": 0.2956,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7817462256315044,
|
| 230 |
+
"grad_norm": 24.18721580505371,
|
| 231 |
+
"learning_rate": 9.806693729891003e-06,
|
| 232 |
+
"loss": 0.2958,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8061757951824888,
|
| 237 |
+
"grad_norm": 34.30314636230469,
|
| 238 |
+
"learning_rate": 9.78070471804521e-06,
|
| 239 |
+
"loss": 0.2972,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8306053647334734,
|
| 244 |
+
"grad_norm": 52.42458724975586,
|
| 245 |
+
"learning_rate": 9.75471570619942e-06,
|
| 246 |
+
"loss": 0.2831,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8550349342844579,
|
| 251 |
+
"grad_norm": 15.497232437133789,
|
| 252 |
+
"learning_rate": 9.728726694353629e-06,
|
| 253 |
+
"loss": 0.2976,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8794645038354424,
|
| 258 |
+
"grad_norm": 30.495269775390625,
|
| 259 |
+
"learning_rate": 9.702737682507836e-06,
|
| 260 |
+
"loss": 0.2977,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9038940733864269,
|
| 265 |
+
"grad_norm": 2.9801318645477295,
|
| 266 |
+
"learning_rate": 9.676748670662046e-06,
|
| 267 |
+
"loss": 0.2834,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.9283236429374114,
|
| 272 |
+
"grad_norm": 12.599749565124512,
|
| 273 |
+
"learning_rate": 9.650759658816253e-06,
|
| 274 |
+
"loss": 0.2919,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.952753212488396,
|
| 279 |
+
"grad_norm": 8.34764575958252,
|
| 280 |
+
"learning_rate": 9.624770646970461e-06,
|
| 281 |
+
"loss": 0.28,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.9771827820393805,
|
| 286 |
+
"grad_norm": 16.70107650756836,
|
| 287 |
+
"learning_rate": 9.59878163512467e-06,
|
| 288 |
+
"loss": 0.2802,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0,
|
| 293 |
+
"eval_accuracy": 0.8791809811735606,
|
| 294 |
+
"eval_averaged_scores": 0.8627373437336334,
|
| 295 |
+
"eval_f1": 0.8462937062937063,
|
| 296 |
+
"eval_loss": 0.2766279876232147,
|
| 297 |
+
"eval_runtime": 20.1271,
|
| 298 |
+
"eval_samples_per_second": 1807.757,
|
| 299 |
+
"eval_steps_per_second": 113.031,
|
| 300 |
+
"step": 20467
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.001612351590365,
|
| 304 |
+
"grad_norm": 2.792717218399048,
|
| 305 |
+
"learning_rate": 9.572792623278878e-06,
|
| 306 |
+
"loss": 0.2782,
|
| 307 |
+
"step": 20500
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.0260419211413494,
|
| 311 |
+
"grad_norm": 8.300065040588379,
|
| 312 |
+
"learning_rate": 9.546803611433086e-06,
|
| 313 |
+
"loss": 0.2515,
|
| 314 |
+
"step": 21000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.050471490692334,
|
| 318 |
+
"grad_norm": 29.82993507385254,
|
| 319 |
+
"learning_rate": 9.520814599587295e-06,
|
| 320 |
+
"loss": 0.2532,
|
| 321 |
+
"step": 21500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.0749010602433184,
|
| 325 |
+
"grad_norm": 51.71006393432617,
|
| 326 |
+
"learning_rate": 9.494825587741504e-06,
|
| 327 |
+
"loss": 0.2411,
|
| 328 |
+
"step": 22000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.099330629794303,
|
| 332 |
+
"grad_norm": 12.74963092803955,
|
| 333 |
+
"learning_rate": 9.468836575895712e-06,
|
| 334 |
+
"loss": 0.2545,
|
| 335 |
+
"step": 22500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.1237601993452875,
|
| 339 |
+
"grad_norm": 11.051137924194336,
|
| 340 |
+
"learning_rate": 9.442847564049921e-06,
|
| 341 |
+
"loss": 0.2597,
|
| 342 |
+
"step": 23000
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.1481897688962721,
|
| 346 |
+
"grad_norm": 16.989593505859375,
|
| 347 |
+
"learning_rate": 9.416858552204129e-06,
|
| 348 |
+
"loss": 0.2508,
|
| 349 |
+
"step": 23500
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.1726193384472565,
|
| 353 |
+
"grad_norm": 18.592727661132812,
|
| 354 |
+
"learning_rate": 9.390869540358337e-06,
|
| 355 |
+
"loss": 0.2518,
|
| 356 |
+
"step": 24000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.1970489079982412,
|
| 360 |
+
"grad_norm": 14.818976402282715,
|
| 361 |
+
"learning_rate": 9.364880528512546e-06,
|
| 362 |
+
"loss": 0.2497,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.2214784775492256,
|
| 367 |
+
"grad_norm": 11.762436866760254,
|
| 368 |
+
"learning_rate": 9.338891516666754e-06,
|
| 369 |
+
"loss": 0.2477,
|
| 370 |
+
"step": 25000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.24590804710021,
|
| 374 |
+
"grad_norm": 3.5736145973205566,
|
| 375 |
+
"learning_rate": 9.312902504820961e-06,
|
| 376 |
+
"loss": 0.2557,
|
| 377 |
+
"step": 25500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.2703376166511946,
|
| 381 |
+
"grad_norm": 17.63505744934082,
|
| 382 |
+
"learning_rate": 9.28691349297517e-06,
|
| 383 |
+
"loss": 0.251,
|
| 384 |
+
"step": 26000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.2947671862021792,
|
| 388 |
+
"grad_norm": 13.431313514709473,
|
| 389 |
+
"learning_rate": 9.26092448112938e-06,
|
| 390 |
+
"loss": 0.2541,
|
| 391 |
+
"step": 26500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.3191967557531636,
|
| 395 |
+
"grad_norm": 29.016206741333008,
|
| 396 |
+
"learning_rate": 9.234935469283588e-06,
|
| 397 |
+
"loss": 0.2417,
|
| 398 |
+
"step": 27000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.343626325304148,
|
| 402 |
+
"grad_norm": 21.277286529541016,
|
| 403 |
+
"learning_rate": 9.208946457437797e-06,
|
| 404 |
+
"loss": 0.2466,
|
| 405 |
+
"step": 27500
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.3680558948551327,
|
| 409 |
+
"grad_norm": 12.927133560180664,
|
| 410 |
+
"learning_rate": 9.182957445592005e-06,
|
| 411 |
+
"loss": 0.2484,
|
| 412 |
+
"step": 28000
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.392485464406117,
|
| 416 |
+
"grad_norm": 14.714786529541016,
|
| 417 |
+
"learning_rate": 9.156968433746212e-06,
|
| 418 |
+
"loss": 0.2543,
|
| 419 |
+
"step": 28500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.4169150339571017,
|
| 423 |
+
"grad_norm": 4.885636806488037,
|
| 424 |
+
"learning_rate": 9.130979421900422e-06,
|
| 425 |
+
"loss": 0.2554,
|
| 426 |
+
"step": 29000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.4413446035080861,
|
| 430 |
+
"grad_norm": 13.911579132080078,
|
| 431 |
+
"learning_rate": 9.10499041005463e-06,
|
| 432 |
+
"loss": 0.2504,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.4657741730590708,
|
| 437 |
+
"grad_norm": 8.80788803100586,
|
| 438 |
+
"learning_rate": 9.079001398208837e-06,
|
| 439 |
+
"loss": 0.2379,
|
| 440 |
+
"step": 30000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.4902037426100552,
|
| 444 |
+
"grad_norm": 14.809807777404785,
|
| 445 |
+
"learning_rate": 9.053012386363046e-06,
|
| 446 |
+
"loss": 0.2462,
|
| 447 |
+
"step": 30500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.5146333121610396,
|
| 451 |
+
"grad_norm": 2.741872549057007,
|
| 452 |
+
"learning_rate": 9.027023374517256e-06,
|
| 453 |
+
"loss": 0.248,
|
| 454 |
+
"step": 31000
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 1.5390628817120242,
|
| 458 |
+
"grad_norm": 15.03188419342041,
|
| 459 |
+
"learning_rate": 9.001034362671463e-06,
|
| 460 |
+
"loss": 0.2523,
|
| 461 |
+
"step": 31500
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.5634924512630088,
|
| 465 |
+
"grad_norm": 13.394209861755371,
|
| 466 |
+
"learning_rate": 8.975045350825672e-06,
|
| 467 |
+
"loss": 0.2445,
|
| 468 |
+
"step": 32000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.5879220208139933,
|
| 472 |
+
"grad_norm": 17.654428482055664,
|
| 473 |
+
"learning_rate": 8.94905633897988e-06,
|
| 474 |
+
"loss": 0.2419,
|
| 475 |
+
"step": 32500
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.6123515903649777,
|
| 479 |
+
"grad_norm": 2.1160449981689453,
|
| 480 |
+
"learning_rate": 8.923067327134088e-06,
|
| 481 |
+
"loss": 0.2423,
|
| 482 |
+
"step": 33000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.6367811599159623,
|
| 486 |
+
"grad_norm": 9.247222900390625,
|
| 487 |
+
"learning_rate": 8.897078315288297e-06,
|
| 488 |
+
"loss": 0.2371,
|
| 489 |
+
"step": 33500
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.661210729466947,
|
| 493 |
+
"grad_norm": 12.957727432250977,
|
| 494 |
+
"learning_rate": 8.871089303442505e-06,
|
| 495 |
+
"loss": 0.245,
|
| 496 |
+
"step": 34000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.6856402990179313,
|
| 500 |
+
"grad_norm": 18.7705020904541,
|
| 501 |
+
"learning_rate": 8.845100291596712e-06,
|
| 502 |
+
"loss": 0.2409,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.7100698685689157,
|
| 507 |
+
"grad_norm": 17.920528411865234,
|
| 508 |
+
"learning_rate": 8.819111279750922e-06,
|
| 509 |
+
"loss": 0.2369,
|
| 510 |
+
"step": 35000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.7344994381199004,
|
| 514 |
+
"grad_norm": 6.508107662200928,
|
| 515 |
+
"learning_rate": 8.793122267905131e-06,
|
| 516 |
+
"loss": 0.2445,
|
| 517 |
+
"step": 35500
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.758929007670885,
|
| 521 |
+
"grad_norm": 15.203210830688477,
|
| 522 |
+
"learning_rate": 8.767133256059339e-06,
|
| 523 |
+
"loss": 0.2441,
|
| 524 |
+
"step": 36000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.7833585772218692,
|
| 528 |
+
"grad_norm": 3.705578327178955,
|
| 529 |
+
"learning_rate": 8.741144244213548e-06,
|
| 530 |
+
"loss": 0.2359,
|
| 531 |
+
"step": 36500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 1.8077881467728538,
|
| 535 |
+
"grad_norm": 26.21458625793457,
|
| 536 |
+
"learning_rate": 8.715155232367756e-06,
|
| 537 |
+
"loss": 0.2339,
|
| 538 |
+
"step": 37000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.8322177163238385,
|
| 542 |
+
"grad_norm": 3.0099897384643555,
|
| 543 |
+
"learning_rate": 8.689166220521963e-06,
|
| 544 |
+
"loss": 0.2376,
|
| 545 |
+
"step": 37500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.8566472858748229,
|
| 549 |
+
"grad_norm": 17.3660888671875,
|
| 550 |
+
"learning_rate": 8.663177208676173e-06,
|
| 551 |
+
"loss": 0.2415,
|
| 552 |
+
"step": 38000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.8810768554258073,
|
| 556 |
+
"grad_norm": 23.040613174438477,
|
| 557 |
+
"learning_rate": 8.63718819683038e-06,
|
| 558 |
+
"loss": 0.236,
|
| 559 |
+
"step": 38500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.905506424976792,
|
| 563 |
+
"grad_norm": 7.123594760894775,
|
| 564 |
+
"learning_rate": 8.61119918498459e-06,
|
| 565 |
+
"loss": 0.2372,
|
| 566 |
+
"step": 39000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.9299359945277765,
|
| 570 |
+
"grad_norm": 34.456966400146484,
|
| 571 |
+
"learning_rate": 8.585210173138797e-06,
|
| 572 |
+
"loss": 0.2391,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.954365564078761,
|
| 577 |
+
"grad_norm": 23.899415969848633,
|
| 578 |
+
"learning_rate": 8.559221161293007e-06,
|
| 579 |
+
"loss": 0.2353,
|
| 580 |
+
"step": 40000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 1.9787951336297454,
|
| 584 |
+
"grad_norm": 19.077669143676758,
|
| 585 |
+
"learning_rate": 8.533232149447214e-06,
|
| 586 |
+
"loss": 0.2369,
|
| 587 |
+
"step": 40500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.0,
|
| 591 |
+
"eval_accuracy": 0.8980623883468463,
|
| 592 |
+
"eval_averaged_scores": 0.8814449118745056,
|
| 593 |
+
"eval_f1": 0.8648274354021648,
|
| 594 |
+
"eval_loss": 0.28552600741386414,
|
| 595 |
+
"eval_runtime": 20.2684,
|
| 596 |
+
"eval_samples_per_second": 1795.163,
|
| 597 |
+
"eval_steps_per_second": 112.244,
|
| 598 |
+
"step": 40934
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.00322470318073,
|
| 602 |
+
"grad_norm": 20.173015594482422,
|
| 603 |
+
"learning_rate": 8.507243137601424e-06,
|
| 604 |
+
"loss": 0.2315,
|
| 605 |
+
"step": 41000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 2.0276542727317146,
|
| 609 |
+
"grad_norm": 39.991127014160156,
|
| 610 |
+
"learning_rate": 8.481254125755631e-06,
|
| 611 |
+
"loss": 0.1966,
|
| 612 |
+
"step": 41500
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.052083842282699,
|
| 616 |
+
"grad_norm": 30.70036506652832,
|
| 617 |
+
"learning_rate": 8.455265113909839e-06,
|
| 618 |
+
"loss": 0.2143,
|
| 619 |
+
"step": 42000
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 2.0765134118336834,
|
| 623 |
+
"grad_norm": 17.682048797607422,
|
| 624 |
+
"learning_rate": 8.429276102064048e-06,
|
| 625 |
+
"loss": 0.2001,
|
| 626 |
+
"step": 42500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.100942981384668,
|
| 630 |
+
"grad_norm": 20.595016479492188,
|
| 631 |
+
"learning_rate": 8.403287090218256e-06,
|
| 632 |
+
"loss": 0.2088,
|
| 633 |
+
"step": 43000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.1253725509356527,
|
| 637 |
+
"grad_norm": 4.142117977142334,
|
| 638 |
+
"learning_rate": 8.377298078372465e-06,
|
| 639 |
+
"loss": 0.2159,
|
| 640 |
+
"step": 43500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.149802120486637,
|
| 644 |
+
"grad_norm": 1.1202857494354248,
|
| 645 |
+
"learning_rate": 8.351309066526673e-06,
|
| 646 |
+
"loss": 0.2031,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 2.1742316900376215,
|
| 651 |
+
"grad_norm": 26.033584594726562,
|
| 652 |
+
"learning_rate": 8.325320054680882e-06,
|
| 653 |
+
"loss": 0.216,
|
| 654 |
+
"step": 44500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.198661259588606,
|
| 658 |
+
"grad_norm": 5.992620468139648,
|
| 659 |
+
"learning_rate": 8.29933104283509e-06,
|
| 660 |
+
"loss": 0.2099,
|
| 661 |
+
"step": 45000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.2230908291395908,
|
| 665 |
+
"grad_norm": 1.6579467058181763,
|
| 666 |
+
"learning_rate": 8.2733420309893e-06,
|
| 667 |
+
"loss": 0.2048,
|
| 668 |
+
"step": 45500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.247520398690575,
|
| 672 |
+
"grad_norm": 30.13793182373047,
|
| 673 |
+
"learning_rate": 8.247353019143507e-06,
|
| 674 |
+
"loss": 0.2148,
|
| 675 |
+
"step": 46000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.2719499682415596,
|
| 679 |
+
"grad_norm": 15.029723167419434,
|
| 680 |
+
"learning_rate": 8.221364007297714e-06,
|
| 681 |
+
"loss": 0.199,
|
| 682 |
+
"step": 46500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.2963795377925442,
|
| 686 |
+
"grad_norm": 7.096563816070557,
|
| 687 |
+
"learning_rate": 8.195374995451924e-06,
|
| 688 |
+
"loss": 0.2133,
|
| 689 |
+
"step": 47000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.3208091073435284,
|
| 693 |
+
"grad_norm": 21.75437355041504,
|
| 694 |
+
"learning_rate": 8.169385983606131e-06,
|
| 695 |
+
"loss": 0.2149,
|
| 696 |
+
"step": 47500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.345238676894513,
|
| 700 |
+
"grad_norm": 17.925037384033203,
|
| 701 |
+
"learning_rate": 8.14339697176034e-06,
|
| 702 |
+
"loss": 0.2006,
|
| 703 |
+
"step": 48000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.3696682464454977,
|
| 707 |
+
"grad_norm": 3.024029493331909,
|
| 708 |
+
"learning_rate": 8.117407959914548e-06,
|
| 709 |
+
"loss": 0.2001,
|
| 710 |
+
"step": 48500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.3940978159964823,
|
| 714 |
+
"grad_norm": 17.959518432617188,
|
| 715 |
+
"learning_rate": 8.091418948068758e-06,
|
| 716 |
+
"loss": 0.2068,
|
| 717 |
+
"step": 49000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.4185273855474665,
|
| 721 |
+
"grad_norm": 12.003142356872559,
|
| 722 |
+
"learning_rate": 8.065429936222965e-06,
|
| 723 |
+
"loss": 0.2113,
|
| 724 |
+
"step": 49500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.442956955098451,
|
| 728 |
+
"grad_norm": 18.196792602539062,
|
| 729 |
+
"learning_rate": 8.039440924377175e-06,
|
| 730 |
+
"loss": 0.2031,
|
| 731 |
+
"step": 50000
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.4673865246494358,
|
| 735 |
+
"grad_norm": 19.81887435913086,
|
| 736 |
+
"learning_rate": 8.013451912531382e-06,
|
| 737 |
+
"loss": 0.2156,
|
| 738 |
+
"step": 50500
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.49181609420042,
|
| 742 |
+
"grad_norm": 20.83788299560547,
|
| 743 |
+
"learning_rate": 7.98746290068559e-06,
|
| 744 |
+
"loss": 0.2036,
|
| 745 |
+
"step": 51000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.5162456637514046,
|
| 749 |
+
"grad_norm": 28.70236587524414,
|
| 750 |
+
"learning_rate": 7.9614738888398e-06,
|
| 751 |
+
"loss": 0.2109,
|
| 752 |
+
"step": 51500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.540675233302389,
|
| 756 |
+
"grad_norm": 28.118547439575195,
|
| 757 |
+
"learning_rate": 7.935484876994007e-06,
|
| 758 |
+
"loss": 0.2022,
|
| 759 |
+
"step": 52000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.565104802853374,
|
| 763 |
+
"grad_norm": 5.818374156951904,
|
| 764 |
+
"learning_rate": 7.909495865148216e-06,
|
| 765 |
+
"loss": 0.2061,
|
| 766 |
+
"step": 52500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.5895343724043585,
|
| 770 |
+
"grad_norm": 11.427398681640625,
|
| 771 |
+
"learning_rate": 7.883506853302424e-06,
|
| 772 |
+
"loss": 0.209,
|
| 773 |
+
"step": 53000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.6139639419553427,
|
| 777 |
+
"grad_norm": 16.268341064453125,
|
| 778 |
+
"learning_rate": 7.857517841456633e-06,
|
| 779 |
+
"loss": 0.2123,
|
| 780 |
+
"step": 53500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.6383935115063273,
|
| 784 |
+
"grad_norm": 23.71199607849121,
|
| 785 |
+
"learning_rate": 7.831528829610841e-06,
|
| 786 |
+
"loss": 0.2125,
|
| 787 |
+
"step": 54000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.662823081057312,
|
| 791 |
+
"grad_norm": 27.364391326904297,
|
| 792 |
+
"learning_rate": 7.80553981776505e-06,
|
| 793 |
+
"loss": 0.2014,
|
| 794 |
+
"step": 54500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.687252650608296,
|
| 798 |
+
"grad_norm": 27.82223129272461,
|
| 799 |
+
"learning_rate": 7.779550805919258e-06,
|
| 800 |
+
"loss": 0.1981,
|
| 801 |
+
"step": 55000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.7116822201592807,
|
| 805 |
+
"grad_norm": 19.91096305847168,
|
| 806 |
+
"learning_rate": 7.753561794073467e-06,
|
| 807 |
+
"loss": 0.2048,
|
| 808 |
+
"step": 55500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.7361117897102654,
|
| 812 |
+
"grad_norm": 22.202478408813477,
|
| 813 |
+
"learning_rate": 7.727572782227675e-06,
|
| 814 |
+
"loss": 0.2077,
|
| 815 |
+
"step": 56000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.76054135926125,
|
| 819 |
+
"grad_norm": 15.403483390808105,
|
| 820 |
+
"learning_rate": 7.701583770381883e-06,
|
| 821 |
+
"loss": 0.2057,
|
| 822 |
+
"step": 56500
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 2.784970928812234,
|
| 826 |
+
"grad_norm": 2.367913007736206,
|
| 827 |
+
"learning_rate": 7.675594758536092e-06,
|
| 828 |
+
"loss": 0.2097,
|
| 829 |
+
"step": 57000
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.809400498363219,
|
| 833 |
+
"grad_norm": 30.33095932006836,
|
| 834 |
+
"learning_rate": 7.6496057466903e-06,
|
| 835 |
+
"loss": 0.2073,
|
| 836 |
+
"step": 57500
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.8338300679142034,
|
| 840 |
+
"grad_norm": 16.65831184387207,
|
| 841 |
+
"learning_rate": 7.623616734844508e-06,
|
| 842 |
+
"loss": 0.2109,
|
| 843 |
+
"step": 58000
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.8582596374651876,
|
| 847 |
+
"grad_norm": 26.70164680480957,
|
| 848 |
+
"learning_rate": 7.597627722998717e-06,
|
| 849 |
+
"loss": 0.1967,
|
| 850 |
+
"step": 58500
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 2.8826892070161723,
|
| 854 |
+
"grad_norm": 15.616362571716309,
|
| 855 |
+
"learning_rate": 7.571638711152926e-06,
|
| 856 |
+
"loss": 0.21,
|
| 857 |
+
"step": 59000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.907118776567157,
|
| 861 |
+
"grad_norm": 1.022119164466858,
|
| 862 |
+
"learning_rate": 7.5456496993071335e-06,
|
| 863 |
+
"loss": 0.2154,
|
| 864 |
+
"step": 59500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.9315483461181415,
|
| 868 |
+
"grad_norm": 23.004011154174805,
|
| 869 |
+
"learning_rate": 7.519660687461342e-06,
|
| 870 |
+
"loss": 0.2081,
|
| 871 |
+
"step": 60000
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 2.955977915669126,
|
| 875 |
+
"grad_norm": 49.54454040527344,
|
| 876 |
+
"learning_rate": 7.4936716756155505e-06,
|
| 877 |
+
"loss": 0.2066,
|
| 878 |
+
"step": 60500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.9804074852201103,
|
| 882 |
+
"grad_norm": 22.06740379333496,
|
| 883 |
+
"learning_rate": 7.467682663769759e-06,
|
| 884 |
+
"loss": 0.2111,
|
| 885 |
+
"step": 61000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.0,
|
| 889 |
+
"eval_accuracy": 0.9057578672529889,
|
| 890 |
+
"eval_averaged_scores": 0.8894141918163889,
|
| 891 |
+
"eval_f1": 0.873070516379789,
|
| 892 |
+
"eval_loss": 0.2923884093761444,
|
| 893 |
+
"eval_runtime": 20.2639,
|
| 894 |
+
"eval_samples_per_second": 1795.555,
|
| 895 |
+
"eval_steps_per_second": 112.268,
|
| 896 |
+
"step": 61401
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 3.004837054771095,
|
| 900 |
+
"grad_norm": 11.454232215881348,
|
| 901 |
+
"learning_rate": 7.441693651923967e-06,
|
| 902 |
+
"loss": 0.2018,
|
| 903 |
+
"step": 61500
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 3.0292666243220796,
|
| 907 |
+
"grad_norm": 21.45491600036621,
|
| 908 |
+
"learning_rate": 7.415704640078175e-06,
|
| 909 |
+
"loss": 0.1833,
|
| 910 |
+
"step": 62000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 3.053696193873064,
|
| 914 |
+
"grad_norm": 0.8129162192344666,
|
| 915 |
+
"learning_rate": 7.389715628232384e-06,
|
| 916 |
+
"loss": 0.1801,
|
| 917 |
+
"step": 62500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.0781257634240484,
|
| 921 |
+
"grad_norm": 0.3495664894580841,
|
| 922 |
+
"learning_rate": 7.363726616386593e-06,
|
| 923 |
+
"loss": 0.1776,
|
| 924 |
+
"step": 63000
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.102555332975033,
|
| 928 |
+
"grad_norm": 0.13657505810260773,
|
| 929 |
+
"learning_rate": 7.337737604540801e-06,
|
| 930 |
+
"loss": 0.182,
|
| 931 |
+
"step": 63500
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 3.1269849025260177,
|
| 935 |
+
"grad_norm": 86.07280731201172,
|
| 936 |
+
"learning_rate": 7.311748592695009e-06,
|
| 937 |
+
"loss": 0.1861,
|
| 938 |
+
"step": 64000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 3.151414472077002,
|
| 942 |
+
"grad_norm": 24.988689422607422,
|
| 943 |
+
"learning_rate": 7.2857595808492175e-06,
|
| 944 |
+
"loss": 0.1776,
|
| 945 |
+
"step": 64500
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 3.1758440416279865,
|
| 949 |
+
"grad_norm": 37.55842208862305,
|
| 950 |
+
"learning_rate": 7.259770569003426e-06,
|
| 951 |
+
"loss": 0.1854,
|
| 952 |
+
"step": 65000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 3.200273611178971,
|
| 956 |
+
"grad_norm": 69.79829406738281,
|
| 957 |
+
"learning_rate": 7.2337815571576345e-06,
|
| 958 |
+
"loss": 0.1747,
|
| 959 |
+
"step": 65500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.2247031807299553,
|
| 963 |
+
"grad_norm": 28.17424201965332,
|
| 964 |
+
"learning_rate": 7.207792545311842e-06,
|
| 965 |
+
"loss": 0.1938,
|
| 966 |
+
"step": 66000
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 3.24913275028094,
|
| 970 |
+
"grad_norm": 56.655364990234375,
|
| 971 |
+
"learning_rate": 7.181803533466051e-06,
|
| 972 |
+
"loss": 0.1912,
|
| 973 |
+
"step": 66500
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 3.2735623198319246,
|
| 977 |
+
"grad_norm": 0.3666965067386627,
|
| 978 |
+
"learning_rate": 7.155814521620259e-06,
|
| 979 |
+
"loss": 0.1915,
|
| 980 |
+
"step": 67000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 3.297991889382909,
|
| 984 |
+
"grad_norm": 9.555418014526367,
|
| 985 |
+
"learning_rate": 7.1298255097744685e-06,
|
| 986 |
+
"loss": 0.188,
|
| 987 |
+
"step": 67500
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 3.3224214589338934,
|
| 991 |
+
"grad_norm": 21.880718231201172,
|
| 992 |
+
"learning_rate": 7.103836497928677e-06,
|
| 993 |
+
"loss": 0.1824,
|
| 994 |
+
"step": 68000
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 3.346851028484878,
|
| 998 |
+
"grad_norm": 34.551021575927734,
|
| 999 |
+
"learning_rate": 7.077847486082885e-06,
|
| 1000 |
+
"loss": 0.1886,
|
| 1001 |
+
"step": 68500
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 3.3712805980358627,
|
| 1005 |
+
"grad_norm": 18.153520584106445,
|
| 1006 |
+
"learning_rate": 7.051858474237093e-06,
|
| 1007 |
+
"loss": 0.1908,
|
| 1008 |
+
"step": 69000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 3.3957101675868473,
|
| 1012 |
+
"grad_norm": 0.3916667103767395,
|
| 1013 |
+
"learning_rate": 7.025869462391302e-06,
|
| 1014 |
+
"loss": 0.182,
|
| 1015 |
+
"step": 69500
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 3.4201397371378315,
|
| 1019 |
+
"grad_norm": 28.306758880615234,
|
| 1020 |
+
"learning_rate": 6.99988045054551e-06,
|
| 1021 |
+
"loss": 0.1852,
|
| 1022 |
+
"step": 70000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.444569306688816,
|
| 1026 |
+
"grad_norm": 29.383926391601562,
|
| 1027 |
+
"learning_rate": 6.973891438699718e-06,
|
| 1028 |
+
"loss": 0.1763,
|
| 1029 |
+
"step": 70500
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 3.4689988762398007,
|
| 1033 |
+
"grad_norm": 16.787750244140625,
|
| 1034 |
+
"learning_rate": 6.947902426853926e-06,
|
| 1035 |
+
"loss": 0.188,
|
| 1036 |
+
"step": 71000
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 3.4934284457907854,
|
| 1040 |
+
"grad_norm": 22.620866775512695,
|
| 1041 |
+
"learning_rate": 6.921913415008135e-06,
|
| 1042 |
+
"loss": 0.193,
|
| 1043 |
+
"step": 71500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.5178580153417696,
|
| 1047 |
+
"grad_norm": 0.9201492667198181,
|
| 1048 |
+
"learning_rate": 6.895924403162344e-06,
|
| 1049 |
+
"loss": 0.186,
|
| 1050 |
+
"step": 72000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 3.542287584892754,
|
| 1054 |
+
"grad_norm": 46.33378601074219,
|
| 1055 |
+
"learning_rate": 6.8699353913165525e-06,
|
| 1056 |
+
"loss": 0.1973,
|
| 1057 |
+
"step": 72500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.566717154443739,
|
| 1061 |
+
"grad_norm": 3.13004207611084,
|
| 1062 |
+
"learning_rate": 6.84394637947076e-06,
|
| 1063 |
+
"loss": 0.2003,
|
| 1064 |
+
"step": 73000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.591146723994723,
|
| 1068 |
+
"grad_norm": 29.041542053222656,
|
| 1069 |
+
"learning_rate": 6.817957367624969e-06,
|
| 1070 |
+
"loss": 0.1874,
|
| 1071 |
+
"step": 73500
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.6155762935457076,
|
| 1075 |
+
"grad_norm": 9.451910018920898,
|
| 1076 |
+
"learning_rate": 6.791968355779177e-06,
|
| 1077 |
+
"loss": 0.1757,
|
| 1078 |
+
"step": 74000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 3.6400058630966923,
|
| 1082 |
+
"grad_norm": 0.4655057191848755,
|
| 1083 |
+
"learning_rate": 6.765979343933386e-06,
|
| 1084 |
+
"loss": 0.1895,
|
| 1085 |
+
"step": 74500
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.664435432647677,
|
| 1089 |
+
"grad_norm": 31.763437271118164,
|
| 1090 |
+
"learning_rate": 6.739990332087593e-06,
|
| 1091 |
+
"loss": 0.1822,
|
| 1092 |
+
"step": 75000
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 3.6888650021986615,
|
| 1096 |
+
"grad_norm": 38.27184295654297,
|
| 1097 |
+
"learning_rate": 6.714001320241802e-06,
|
| 1098 |
+
"loss": 0.1849,
|
| 1099 |
+
"step": 75500
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 3.7132945717496457,
|
| 1103 |
+
"grad_norm": 6.707937717437744,
|
| 1104 |
+
"learning_rate": 6.68801230839601e-06,
|
| 1105 |
+
"loss": 0.2007,
|
| 1106 |
+
"step": 76000
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.7377241413006304,
|
| 1110 |
+
"grad_norm": 30.213747024536133,
|
| 1111 |
+
"learning_rate": 6.66202329655022e-06,
|
| 1112 |
+
"loss": 0.1903,
|
| 1113 |
+
"step": 76500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 3.7621537108516145,
|
| 1117 |
+
"grad_norm": 51.97517776489258,
|
| 1118 |
+
"learning_rate": 6.636034284704428e-06,
|
| 1119 |
+
"loss": 0.1875,
|
| 1120 |
+
"step": 77000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 3.786583280402599,
|
| 1124 |
+
"grad_norm": 28.538976669311523,
|
| 1125 |
+
"learning_rate": 6.610045272858637e-06,
|
| 1126 |
+
"loss": 0.1857,
|
| 1127 |
+
"step": 77500
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 3.811012849953584,
|
| 1131 |
+
"grad_norm": 0.3994407057762146,
|
| 1132 |
+
"learning_rate": 6.584056261012844e-06,
|
| 1133 |
+
"loss": 0.1819,
|
| 1134 |
+
"step": 78000
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 3.8354424195045684,
|
| 1138 |
+
"grad_norm": 28.58909797668457,
|
| 1139 |
+
"learning_rate": 6.558067249167053e-06,
|
| 1140 |
+
"loss": 0.1787,
|
| 1141 |
+
"step": 78500
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 3.859871989055553,
|
| 1145 |
+
"grad_norm": 0.4431411325931549,
|
| 1146 |
+
"learning_rate": 6.532078237321261e-06,
|
| 1147 |
+
"loss": 0.1864,
|
| 1148 |
+
"step": 79000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 3.8843015586065373,
|
| 1152 |
+
"grad_norm": 65.34574890136719,
|
| 1153 |
+
"learning_rate": 6.50608922547547e-06,
|
| 1154 |
+
"loss": 0.1844,
|
| 1155 |
+
"step": 79500
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 3.908731128157522,
|
| 1159 |
+
"grad_norm": 12.750852584838867,
|
| 1160 |
+
"learning_rate": 6.480100213629677e-06,
|
| 1161 |
+
"loss": 0.1824,
|
| 1162 |
+
"step": 80000
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 3.9331606977085065,
|
| 1166 |
+
"grad_norm": 39.058040618896484,
|
| 1167 |
+
"learning_rate": 6.454111201783886e-06,
|
| 1168 |
+
"loss": 0.1849,
|
| 1169 |
+
"step": 80500
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 3.9575902672594907,
|
| 1173 |
+
"grad_norm": 2.713413715362549,
|
| 1174 |
+
"learning_rate": 6.428122189938095e-06,
|
| 1175 |
+
"loss": 0.1768,
|
| 1176 |
+
"step": 81000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 3.9820198368104753,
|
| 1180 |
+
"grad_norm": 1.6995596885681152,
|
| 1181 |
+
"learning_rate": 6.402133178092304e-06,
|
| 1182 |
+
"loss": 0.1835,
|
| 1183 |
+
"step": 81500
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 4.0,
|
| 1187 |
+
"eval_accuracy": 0.9077641885392332,
|
| 1188 |
+
"eval_averaged_scores": 0.8907519212071636,
|
| 1189 |
+
"eval_f1": 0.873739653875094,
|
| 1190 |
+
"eval_loss": 0.33108219504356384,
|
| 1191 |
+
"eval_runtime": 20.2844,
|
| 1192 |
+
"eval_samples_per_second": 1793.747,
|
| 1193 |
+
"eval_steps_per_second": 112.155,
|
| 1194 |
+
"step": 81868
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 4.00644940636146,
|
| 1198 |
+
"grad_norm": 0.7421801686286926,
|
| 1199 |
+
"learning_rate": 6.376144166246512e-06,
|
| 1200 |
+
"loss": 0.1796,
|
| 1201 |
+
"step": 82000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 4.030878975912445,
|
| 1205 |
+
"grad_norm": 30.367263793945312,
|
| 1206 |
+
"learning_rate": 6.35015515440072e-06,
|
| 1207 |
+
"loss": 0.1649,
|
| 1208 |
+
"step": 82500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 4.055308545463429,
|
| 1212 |
+
"grad_norm": 14.93786907196045,
|
| 1213 |
+
"learning_rate": 6.324166142554928e-06,
|
| 1214 |
+
"loss": 0.164,
|
| 1215 |
+
"step": 83000
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 4.079738115014414,
|
| 1219 |
+
"grad_norm": 0.6399831175804138,
|
| 1220 |
+
"learning_rate": 6.298177130709137e-06,
|
| 1221 |
+
"loss": 0.1605,
|
| 1222 |
+
"step": 83500
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 4.104167684565398,
|
| 1226 |
+
"grad_norm": 23.02483558654785,
|
| 1227 |
+
"learning_rate": 6.272188118863345e-06,
|
| 1228 |
+
"loss": 0.1659,
|
| 1229 |
+
"step": 84000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 4.128597254116382,
|
| 1233 |
+
"grad_norm": 0.35685986280441284,
|
| 1234 |
+
"learning_rate": 6.246199107017553e-06,
|
| 1235 |
+
"loss": 0.1573,
|
| 1236 |
+
"step": 84500
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 4.153026823667367,
|
| 1240 |
+
"grad_norm": 52.071285247802734,
|
| 1241 |
+
"learning_rate": 6.220210095171761e-06,
|
| 1242 |
+
"loss": 0.1683,
|
| 1243 |
+
"step": 85000
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 4.1774563932183515,
|
| 1247 |
+
"grad_norm": 4.686119556427002,
|
| 1248 |
+
"learning_rate": 6.194221083325971e-06,
|
| 1249 |
+
"loss": 0.1641,
|
| 1250 |
+
"step": 85500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 4.201885962769336,
|
| 1254 |
+
"grad_norm": 0.708002507686615,
|
| 1255 |
+
"learning_rate": 6.168232071480179e-06,
|
| 1256 |
+
"loss": 0.1603,
|
| 1257 |
+
"step": 86000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 4.226315532320321,
|
| 1261 |
+
"grad_norm": 1.475074291229248,
|
| 1262 |
+
"learning_rate": 6.142243059634388e-06,
|
| 1263 |
+
"loss": 0.1671,
|
| 1264 |
+
"step": 86500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 4.250745101871305,
|
| 1268 |
+
"grad_norm": 1.395318627357483,
|
| 1269 |
+
"learning_rate": 6.116254047788595e-06,
|
| 1270 |
+
"loss": 0.1575,
|
| 1271 |
+
"step": 87000
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 4.275174671422289,
|
| 1275 |
+
"grad_norm": 0.3956087827682495,
|
| 1276 |
+
"learning_rate": 6.090265035942804e-06,
|
| 1277 |
+
"loss": 0.1591,
|
| 1278 |
+
"step": 87500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 4.299604240973274,
|
| 1282 |
+
"grad_norm": 0.6056288480758667,
|
| 1283 |
+
"learning_rate": 6.064276024097012e-06,
|
| 1284 |
+
"loss": 0.1614,
|
| 1285 |
+
"step": 88000
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 4.324033810524258,
|
| 1289 |
+
"grad_norm": 1.0418305397033691,
|
| 1290 |
+
"learning_rate": 6.038287012251221e-06,
|
| 1291 |
+
"loss": 0.1715,
|
| 1292 |
+
"step": 88500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 4.348463380075243,
|
| 1296 |
+
"grad_norm": 1.0520128011703491,
|
| 1297 |
+
"learning_rate": 6.0122980004054285e-06,
|
| 1298 |
+
"loss": 0.1762,
|
| 1299 |
+
"step": 89000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 4.372892949626228,
|
| 1303 |
+
"grad_norm": 0.8157211542129517,
|
| 1304 |
+
"learning_rate": 5.986308988559637e-06,
|
| 1305 |
+
"loss": 0.1702,
|
| 1306 |
+
"step": 89500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 4.397322519177212,
|
| 1310 |
+
"grad_norm": 21.881351470947266,
|
| 1311 |
+
"learning_rate": 5.9603199767138455e-06,
|
| 1312 |
+
"loss": 0.1649,
|
| 1313 |
+
"step": 90000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 4.421752088728197,
|
| 1317 |
+
"grad_norm": 21.02885627746582,
|
| 1318 |
+
"learning_rate": 5.934330964868055e-06,
|
| 1319 |
+
"loss": 0.1681,
|
| 1320 |
+
"step": 90500
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 4.4461816582791815,
|
| 1324 |
+
"grad_norm": 17.863567352294922,
|
| 1325 |
+
"learning_rate": 5.908341953022263e-06,
|
| 1326 |
+
"loss": 0.1705,
|
| 1327 |
+
"step": 91000
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 4.470611227830165,
|
| 1331 |
+
"grad_norm": 66.63270568847656,
|
| 1332 |
+
"learning_rate": 5.882352941176471e-06,
|
| 1333 |
+
"loss": 0.1674,
|
| 1334 |
+
"step": 91500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 4.49504079738115,
|
| 1338 |
+
"grad_norm": 14.567200660705566,
|
| 1339 |
+
"learning_rate": 5.856363929330679e-06,
|
| 1340 |
+
"loss": 0.1612,
|
| 1341 |
+
"step": 92000
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 4.5194703669321346,
|
| 1345 |
+
"grad_norm": 2.0841243267059326,
|
| 1346 |
+
"learning_rate": 5.830374917484888e-06,
|
| 1347 |
+
"loss": 0.1588,
|
| 1348 |
+
"step": 92500
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 4.543899936483119,
|
| 1352 |
+
"grad_norm": 21.195947647094727,
|
| 1353 |
+
"learning_rate": 5.804385905639096e-06,
|
| 1354 |
+
"loss": 0.1686,
|
| 1355 |
+
"step": 93000
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 4.568329506034104,
|
| 1359 |
+
"grad_norm": 19.75173568725586,
|
| 1360 |
+
"learning_rate": 5.778396893793304e-06,
|
| 1361 |
+
"loss": 0.1718,
|
| 1362 |
+
"step": 93500
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 4.5927590755850884,
|
| 1366 |
+
"grad_norm": 34.48356628417969,
|
| 1367 |
+
"learning_rate": 5.7524078819475125e-06,
|
| 1368 |
+
"loss": 0.1693,
|
| 1369 |
+
"step": 94000
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 4.617188645136073,
|
| 1373 |
+
"grad_norm": 66.5281982421875,
|
| 1374 |
+
"learning_rate": 5.726418870101721e-06,
|
| 1375 |
+
"loss": 0.1654,
|
| 1376 |
+
"step": 94500
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 4.641618214687057,
|
| 1380 |
+
"grad_norm": 21.60127830505371,
|
| 1381 |
+
"learning_rate": 5.70042985825593e-06,
|
| 1382 |
+
"loss": 0.1525,
|
| 1383 |
+
"step": 95000
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 4.6660477842380415,
|
| 1387 |
+
"grad_norm": 53.85319900512695,
|
| 1388 |
+
"learning_rate": 5.674440846410139e-06,
|
| 1389 |
+
"loss": 0.1683,
|
| 1390 |
+
"step": 95500
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 4.690477353789026,
|
| 1394 |
+
"grad_norm": 21.81485366821289,
|
| 1395 |
+
"learning_rate": 5.6484518345643465e-06,
|
| 1396 |
+
"loss": 0.1597,
|
| 1397 |
+
"step": 96000
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 4.714906923340011,
|
| 1401 |
+
"grad_norm": 3.4359285831451416,
|
| 1402 |
+
"learning_rate": 5.622462822718555e-06,
|
| 1403 |
+
"loss": 0.1744,
|
| 1404 |
+
"step": 96500
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 4.739336492890995,
|
| 1408 |
+
"grad_norm": 37.74496078491211,
|
| 1409 |
+
"learning_rate": 5.5964738108727635e-06,
|
| 1410 |
+
"loss": 0.1549,
|
| 1411 |
+
"step": 97000
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 4.76376606244198,
|
| 1415 |
+
"grad_norm": 46.64603042602539,
|
| 1416 |
+
"learning_rate": 5.570484799026972e-06,
|
| 1417 |
+
"loss": 0.1628,
|
| 1418 |
+
"step": 97500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 4.788195631992965,
|
| 1422 |
+
"grad_norm": 1.0236244201660156,
|
| 1423 |
+
"learning_rate": 5.54449578718118e-06,
|
| 1424 |
+
"loss": 0.1543,
|
| 1425 |
+
"step": 98000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 4.812625201543948,
|
| 1429 |
+
"grad_norm": 75.760009765625,
|
| 1430 |
+
"learning_rate": 5.518506775335388e-06,
|
| 1431 |
+
"loss": 0.1788,
|
| 1432 |
+
"step": 98500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 4.837054771094933,
|
| 1436 |
+
"grad_norm": 1.3948771953582764,
|
| 1437 |
+
"learning_rate": 5.4925177634895966e-06,
|
| 1438 |
+
"loss": 0.1675,
|
| 1439 |
+
"step": 99000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 4.861484340645918,
|
| 1443 |
+
"grad_norm": 25.799287796020508,
|
| 1444 |
+
"learning_rate": 5.466528751643806e-06,
|
| 1445 |
+
"loss": 0.1756,
|
| 1446 |
+
"step": 99500
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 4.885913910196902,
|
| 1450 |
+
"grad_norm": 10.39145278930664,
|
| 1451 |
+
"learning_rate": 5.440539739798014e-06,
|
| 1452 |
+
"loss": 0.168,
|
| 1453 |
+
"step": 100000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 4.910343479747887,
|
| 1457 |
+
"grad_norm": 0.5542752742767334,
|
| 1458 |
+
"learning_rate": 5.414550727952223e-06,
|
| 1459 |
+
"loss": 0.158,
|
| 1460 |
+
"step": 100500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 4.9347730492988715,
|
| 1464 |
+
"grad_norm": 2.8913373947143555,
|
| 1465 |
+
"learning_rate": 5.3885617161064305e-06,
|
| 1466 |
+
"loss": 0.1703,
|
| 1467 |
+
"step": 101000
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 4.959202618849856,
|
| 1471 |
+
"grad_norm": 76.17244720458984,
|
| 1472 |
+
"learning_rate": 5.362572704260639e-06,
|
| 1473 |
+
"loss": 0.1571,
|
| 1474 |
+
"step": 101500
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 4.98363218840084,
|
| 1478 |
+
"grad_norm": 29.392118453979492,
|
| 1479 |
+
"learning_rate": 5.3365836924148475e-06,
|
| 1480 |
+
"loss": 0.1671,
|
| 1481 |
+
"step": 102000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 5.0,
|
| 1485 |
+
"eval_accuracy": 0.9093857358801704,
|
| 1486 |
+
"eval_averaged_scores": 0.893972399414541,
|
| 1487 |
+
"eval_f1": 0.8785590629489115,
|
| 1488 |
+
"eval_loss": 0.37774011492729187,
|
| 1489 |
+
"eval_runtime": 20.1329,
|
| 1490 |
+
"eval_samples_per_second": 1807.24,
|
| 1491 |
+
"eval_steps_per_second": 112.999,
|
| 1492 |
+
"step": 102335
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 5.0080617579518245,
|
| 1496 |
+
"grad_norm": 0.33907392621040344,
|
| 1497 |
+
"learning_rate": 5.310594680569056e-06,
|
| 1498 |
+
"loss": 0.1594,
|
| 1499 |
+
"step": 102500
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 5.032491327502809,
|
| 1503 |
+
"grad_norm": 0.20179282128810883,
|
| 1504 |
+
"learning_rate": 5.284605668723264e-06,
|
| 1505 |
+
"loss": 0.1234,
|
| 1506 |
+
"step": 103000
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 5.056920897053794,
|
| 1510 |
+
"grad_norm": 0.2239135205745697,
|
| 1511 |
+
"learning_rate": 5.258616656877472e-06,
|
| 1512 |
+
"loss": 0.1398,
|
| 1513 |
+
"step": 103500
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 5.081350466604778,
|
| 1517 |
+
"grad_norm": 0.2501750886440277,
|
| 1518 |
+
"learning_rate": 5.2326276450316815e-06,
|
| 1519 |
+
"loss": 0.141,
|
| 1520 |
+
"step": 104000
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 5.105780036155763,
|
| 1524 |
+
"grad_norm": 27.644052505493164,
|
| 1525 |
+
"learning_rate": 5.20663863318589e-06,
|
| 1526 |
+
"loss": 0.1447,
|
| 1527 |
+
"step": 104500
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 5.130209605706748,
|
| 1531 |
+
"grad_norm": 0.7906270623207092,
|
| 1532 |
+
"learning_rate": 5.1806496213400985e-06,
|
| 1533 |
+
"loss": 0.1368,
|
| 1534 |
+
"step": 105000
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 5.154639175257732,
|
| 1538 |
+
"grad_norm": 0.17316019535064697,
|
| 1539 |
+
"learning_rate": 5.154660609494306e-06,
|
| 1540 |
+
"loss": 0.131,
|
| 1541 |
+
"step": 105500
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 5.179068744808716,
|
| 1545 |
+
"grad_norm": 15.188929557800293,
|
| 1546 |
+
"learning_rate": 5.128671597648515e-06,
|
| 1547 |
+
"loss": 0.1351,
|
| 1548 |
+
"step": 106000
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 5.203498314359701,
|
| 1552 |
+
"grad_norm": 0.39822831749916077,
|
| 1553 |
+
"learning_rate": 5.102682585802723e-06,
|
| 1554 |
+
"loss": 0.1424,
|
| 1555 |
+
"step": 106500
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 5.227927883910685,
|
| 1559 |
+
"grad_norm": 0.22537678480148315,
|
| 1560 |
+
"learning_rate": 5.0766935739569316e-06,
|
| 1561 |
+
"loss": 0.1369,
|
| 1562 |
+
"step": 107000
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 5.25235745346167,
|
| 1566 |
+
"grad_norm": 1.028478741645813,
|
| 1567 |
+
"learning_rate": 5.050704562111139e-06,
|
| 1568 |
+
"loss": 0.1358,
|
| 1569 |
+
"step": 107500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 5.276787023012655,
|
| 1573 |
+
"grad_norm": 23.18048858642578,
|
| 1574 |
+
"learning_rate": 5.024715550265348e-06,
|
| 1575 |
+
"loss": 0.1335,
|
| 1576 |
+
"step": 108000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 5.301216592563639,
|
| 1580 |
+
"grad_norm": 22.66480255126953,
|
| 1581 |
+
"learning_rate": 4.998726538419556e-06,
|
| 1582 |
+
"loss": 0.1358,
|
| 1583 |
+
"step": 108500
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 5.325646162114624,
|
| 1587 |
+
"grad_norm": 1.0610641241073608,
|
| 1588 |
+
"learning_rate": 4.972737526573765e-06,
|
| 1589 |
+
"loss": 0.1385,
|
| 1590 |
+
"step": 109000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 5.3500757316656085,
|
| 1594 |
+
"grad_norm": 0.12008700519800186,
|
| 1595 |
+
"learning_rate": 4.946748514727974e-06,
|
| 1596 |
+
"loss": 0.1462,
|
| 1597 |
+
"step": 109500
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 5.374505301216592,
|
| 1601 |
+
"grad_norm": 0.47721561789512634,
|
| 1602 |
+
"learning_rate": 4.920759502882182e-06,
|
| 1603 |
+
"loss": 0.1407,
|
| 1604 |
+
"step": 110000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 5.398934870767577,
|
| 1608 |
+
"grad_norm": 0.9414154291152954,
|
| 1609 |
+
"learning_rate": 4.89477049103639e-06,
|
| 1610 |
+
"loss": 0.1435,
|
| 1611 |
+
"step": 110500
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 5.4233644403185615,
|
| 1615 |
+
"grad_norm": 260.96466064453125,
|
| 1616 |
+
"learning_rate": 4.868781479190599e-06,
|
| 1617 |
+
"loss": 0.1319,
|
| 1618 |
+
"step": 111000
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 5.447794009869546,
|
| 1622 |
+
"grad_norm": 0.2496846616268158,
|
| 1623 |
+
"learning_rate": 4.842792467344807e-06,
|
| 1624 |
+
"loss": 0.1544,
|
| 1625 |
+
"step": 111500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 5.472223579420531,
|
| 1629 |
+
"grad_norm": 57.217315673828125,
|
| 1630 |
+
"learning_rate": 4.816803455499016e-06,
|
| 1631 |
+
"loss": 0.1328,
|
| 1632 |
+
"step": 112000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 5.496653148971515,
|
| 1636 |
+
"grad_norm": 0.7028847932815552,
|
| 1637 |
+
"learning_rate": 4.790814443653224e-06,
|
| 1638 |
+
"loss": 0.1396,
|
| 1639 |
+
"step": 112500
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 5.5210827185225,
|
| 1643 |
+
"grad_norm": 0.15381655097007751,
|
| 1644 |
+
"learning_rate": 4.764825431807433e-06,
|
| 1645 |
+
"loss": 0.1384,
|
| 1646 |
+
"step": 113000
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 5.545512288073484,
|
| 1650 |
+
"grad_norm": 1.1577835083007812,
|
| 1651 |
+
"learning_rate": 4.73883641996164e-06,
|
| 1652 |
+
"loss": 0.1447,
|
| 1653 |
+
"step": 113500
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 5.569941857624468,
|
| 1657 |
+
"grad_norm": 0.1358271986246109,
|
| 1658 |
+
"learning_rate": 4.712847408115849e-06,
|
| 1659 |
+
"loss": 0.1311,
|
| 1660 |
+
"step": 114000
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 5.594371427175453,
|
| 1664 |
+
"grad_norm": 44.117523193359375,
|
| 1665 |
+
"learning_rate": 4.686858396270057e-06,
|
| 1666 |
+
"loss": 0.1533,
|
| 1667 |
+
"step": 114500
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 5.618800996726438,
|
| 1671 |
+
"grad_norm": 1.6837427616119385,
|
| 1672 |
+
"learning_rate": 4.660869384424266e-06,
|
| 1673 |
+
"loss": 0.1392,
|
| 1674 |
+
"step": 115000
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 5.643230566277422,
|
| 1678 |
+
"grad_norm": 0.5804646611213684,
|
| 1679 |
+
"learning_rate": 4.634880372578474e-06,
|
| 1680 |
+
"loss": 0.1416,
|
| 1681 |
+
"step": 115500
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 5.667660135828407,
|
| 1685 |
+
"grad_norm": 6.186709880828857,
|
| 1686 |
+
"learning_rate": 4.608891360732683e-06,
|
| 1687 |
+
"loss": 0.1377,
|
| 1688 |
+
"step": 116000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 5.6920897053793915,
|
| 1692 |
+
"grad_norm": 3.6101438999176025,
|
| 1693 |
+
"learning_rate": 4.582902348886891e-06,
|
| 1694 |
+
"loss": 0.1355,
|
| 1695 |
+
"step": 116500
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 5.716519274930375,
|
| 1699 |
+
"grad_norm": 59.04292297363281,
|
| 1700 |
+
"learning_rate": 4.5569133370411e-06,
|
| 1701 |
+
"loss": 0.1418,
|
| 1702 |
+
"step": 117000
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 5.74094884448136,
|
| 1706 |
+
"grad_norm": 103.19622802734375,
|
| 1707 |
+
"learning_rate": 4.530924325195308e-06,
|
| 1708 |
+
"loss": 0.1385,
|
| 1709 |
+
"step": 117500
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 5.7653784140323445,
|
| 1713 |
+
"grad_norm": 0.046471111476421356,
|
| 1714 |
+
"learning_rate": 4.504935313349516e-06,
|
| 1715 |
+
"loss": 0.1437,
|
| 1716 |
+
"step": 118000
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 5.789807983583329,
|
| 1720 |
+
"grad_norm": 14.723210334777832,
|
| 1721 |
+
"learning_rate": 4.478946301503724e-06,
|
| 1722 |
+
"loss": 0.1489,
|
| 1723 |
+
"step": 118500
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 5.814237553134314,
|
| 1727 |
+
"grad_norm": 0.23389513790607452,
|
| 1728 |
+
"learning_rate": 4.452957289657933e-06,
|
| 1729 |
+
"loss": 0.1368,
|
| 1730 |
+
"step": 119000
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 5.838667122685298,
|
| 1734 |
+
"grad_norm": 52.34353256225586,
|
| 1735 |
+
"learning_rate": 4.426968277812141e-06,
|
| 1736 |
+
"loss": 0.1366,
|
| 1737 |
+
"step": 119500
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 5.863096692236283,
|
| 1741 |
+
"grad_norm": 0.24756839871406555,
|
| 1742 |
+
"learning_rate": 4.40097926596635e-06,
|
| 1743 |
+
"loss": 0.1466,
|
| 1744 |
+
"step": 120000
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 5.887526261787268,
|
| 1748 |
+
"grad_norm": 12.284710884094238,
|
| 1749 |
+
"learning_rate": 4.374990254120558e-06,
|
| 1750 |
+
"loss": 0.1477,
|
| 1751 |
+
"step": 120500
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 5.911955831338252,
|
| 1755 |
+
"grad_norm": 4.85374116897583,
|
| 1756 |
+
"learning_rate": 4.349001242274767e-06,
|
| 1757 |
+
"loss": 0.1406,
|
| 1758 |
+
"step": 121000
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 5.936385400889236,
|
| 1762 |
+
"grad_norm": 23.76519012451172,
|
| 1763 |
+
"learning_rate": 4.323012230428975e-06,
|
| 1764 |
+
"loss": 0.1434,
|
| 1765 |
+
"step": 121500
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 5.960814970440221,
|
| 1769 |
+
"grad_norm": 0.7099828720092773,
|
| 1770 |
+
"learning_rate": 4.297023218583184e-06,
|
| 1771 |
+
"loss": 0.1524,
|
| 1772 |
+
"step": 122000
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 5.985244539991205,
|
| 1776 |
+
"grad_norm": 30.665578842163086,
|
| 1777 |
+
"learning_rate": 4.271034206737391e-06,
|
| 1778 |
+
"loss": 0.1381,
|
| 1779 |
+
"step": 122500
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 6.0,
|
| 1783 |
+
"eval_accuracy": 0.9105950254225642,
|
| 1784 |
+
"eval_averaged_scores": 0.8952502575704396,
|
| 1785 |
+
"eval_f1": 0.8799054897183151,
|
| 1786 |
+
"eval_loss": 0.44903886318206787,
|
| 1787 |
+
"eval_runtime": 20.1055,
|
| 1788 |
+
"eval_samples_per_second": 1809.706,
|
| 1789 |
+
"eval_steps_per_second": 113.153,
|
| 1790 |
+
"step": 122802
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 6.00967410954219,
|
| 1794 |
+
"grad_norm": 21.405393600463867,
|
| 1795 |
+
"learning_rate": 4.2450451948916e-06,
|
| 1796 |
+
"loss": 0.1166,
|
| 1797 |
+
"step": 123000
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 6.034103679093175,
|
| 1801 |
+
"grad_norm": 8.164361953735352,
|
| 1802 |
+
"learning_rate": 4.219056183045809e-06,
|
| 1803 |
+
"loss": 0.1086,
|
| 1804 |
+
"step": 123500
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 6.058533248644159,
|
| 1808 |
+
"grad_norm": 0.2232547402381897,
|
| 1809 |
+
"learning_rate": 4.193067171200017e-06,
|
| 1810 |
+
"loss": 0.1152,
|
| 1811 |
+
"step": 124000
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 6.082962818195144,
|
| 1815 |
+
"grad_norm": 72.73697662353516,
|
| 1816 |
+
"learning_rate": 4.167078159354225e-06,
|
| 1817 |
+
"loss": 0.109,
|
| 1818 |
+
"step": 124500
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 6.107392387746128,
|
| 1822 |
+
"grad_norm": 0.5123595595359802,
|
| 1823 |
+
"learning_rate": 4.141089147508434e-06,
|
| 1824 |
+
"loss": 0.1183,
|
| 1825 |
+
"step": 125000
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 6.131821957297112,
|
| 1829 |
+
"grad_norm": 0.04295356571674347,
|
| 1830 |
+
"learning_rate": 4.115100135662642e-06,
|
| 1831 |
+
"loss": 0.1102,
|
| 1832 |
+
"step": 125500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 6.156251526848097,
|
| 1836 |
+
"grad_norm": 17.82163429260254,
|
| 1837 |
+
"learning_rate": 4.089111123816851e-06,
|
| 1838 |
+
"loss": 0.1059,
|
| 1839 |
+
"step": 126000
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 6.1806810963990815,
|
| 1843 |
+
"grad_norm": 0.28495362401008606,
|
| 1844 |
+
"learning_rate": 4.063122111971059e-06,
|
| 1845 |
+
"loss": 0.1053,
|
| 1846 |
+
"step": 126500
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 6.205110665950066,
|
| 1850 |
+
"grad_norm": 9.597000122070312,
|
| 1851 |
+
"learning_rate": 4.037133100125267e-06,
|
| 1852 |
+
"loss": 0.1244,
|
| 1853 |
+
"step": 127000
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 6.229540235501051,
|
| 1857 |
+
"grad_norm": 0.0872587263584137,
|
| 1858 |
+
"learning_rate": 4.011144088279475e-06,
|
| 1859 |
+
"loss": 0.1191,
|
| 1860 |
+
"step": 127500
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 6.253969805052035,
|
| 1864 |
+
"grad_norm": 22.000926971435547,
|
| 1865 |
+
"learning_rate": 3.985155076433685e-06,
|
| 1866 |
+
"loss": 0.1243,
|
| 1867 |
+
"step": 128000
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 6.278399374603019,
|
| 1871 |
+
"grad_norm": 9.766866683959961,
|
| 1872 |
+
"learning_rate": 3.959166064587892e-06,
|
| 1873 |
+
"loss": 0.1202,
|
| 1874 |
+
"step": 128500
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 6.302828944154004,
|
| 1878 |
+
"grad_norm": 12.51276969909668,
|
| 1879 |
+
"learning_rate": 3.933177052742101e-06,
|
| 1880 |
+
"loss": 0.1167,
|
| 1881 |
+
"step": 129000
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 6.327258513704988,
|
| 1885 |
+
"grad_norm": 0.99669349193573,
|
| 1886 |
+
"learning_rate": 3.907188040896309e-06,
|
| 1887 |
+
"loss": 0.1157,
|
| 1888 |
+
"step": 129500
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 6.351688083255973,
|
| 1892 |
+
"grad_norm": 0.4419931471347809,
|
| 1893 |
+
"learning_rate": 3.881199029050518e-06,
|
| 1894 |
+
"loss": 0.113,
|
| 1895 |
+
"step": 130000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 6.376117652806958,
|
| 1899 |
+
"grad_norm": 0.0723571702837944,
|
| 1900 |
+
"learning_rate": 3.855210017204726e-06,
|
| 1901 |
+
"loss": 0.1239,
|
| 1902 |
+
"step": 130500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 6.400547222357942,
|
| 1906 |
+
"grad_norm": 67.760009765625,
|
| 1907 |
+
"learning_rate": 3.829221005358935e-06,
|
| 1908 |
+
"loss": 0.1206,
|
| 1909 |
+
"step": 131000
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 6.424976791908927,
|
| 1913 |
+
"grad_norm": 145.9121551513672,
|
| 1914 |
+
"learning_rate": 3.803231993513143e-06,
|
| 1915 |
+
"loss": 0.1153,
|
| 1916 |
+
"step": 131500
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 6.449406361459911,
|
| 1920 |
+
"grad_norm": 14.276485443115234,
|
| 1921 |
+
"learning_rate": 3.777242981667351e-06,
|
| 1922 |
+
"loss": 0.1255,
|
| 1923 |
+
"step": 132000
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 6.473835931010895,
|
| 1927 |
+
"grad_norm": 0.9386702179908752,
|
| 1928 |
+
"learning_rate": 3.75125396982156e-06,
|
| 1929 |
+
"loss": 0.1196,
|
| 1930 |
+
"step": 132500
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 6.49826550056188,
|
| 1934 |
+
"grad_norm": 0.18407511711120605,
|
| 1935 |
+
"learning_rate": 3.7252649579757684e-06,
|
| 1936 |
+
"loss": 0.1138,
|
| 1937 |
+
"step": 133000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 6.5226950701128645,
|
| 1941 |
+
"grad_norm": 0.8368340730667114,
|
| 1942 |
+
"learning_rate": 3.6992759461299764e-06,
|
| 1943 |
+
"loss": 0.1273,
|
| 1944 |
+
"step": 133500
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 6.547124639663849,
|
| 1948 |
+
"grad_norm": 8.838400840759277,
|
| 1949 |
+
"learning_rate": 3.673286934284185e-06,
|
| 1950 |
+
"loss": 0.1103,
|
| 1951 |
+
"step": 134000
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 6.571554209214834,
|
| 1955 |
+
"grad_norm": 0.08394081890583038,
|
| 1956 |
+
"learning_rate": 3.647297922438393e-06,
|
| 1957 |
+
"loss": 0.1205,
|
| 1958 |
+
"step": 134500
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 6.595983778765818,
|
| 1962 |
+
"grad_norm": 0.11461492627859116,
|
| 1963 |
+
"learning_rate": 3.621308910592602e-06,
|
| 1964 |
+
"loss": 0.1103,
|
| 1965 |
+
"step": 135000
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 6.620413348316802,
|
| 1969 |
+
"grad_norm": 0.16645289957523346,
|
| 1970 |
+
"learning_rate": 3.5953198987468104e-06,
|
| 1971 |
+
"loss": 0.1154,
|
| 1972 |
+
"step": 135500
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 6.644842917867787,
|
| 1976 |
+
"grad_norm": 288.46868896484375,
|
| 1977 |
+
"learning_rate": 3.5693308869010185e-06,
|
| 1978 |
+
"loss": 0.1281,
|
| 1979 |
+
"step": 136000
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 6.669272487418771,
|
| 1983 |
+
"grad_norm": 0.03742976859211922,
|
| 1984 |
+
"learning_rate": 3.543341875055227e-06,
|
| 1985 |
+
"loss": 0.1225,
|
| 1986 |
+
"step": 136500
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 6.693702056969756,
|
| 1990 |
+
"grad_norm": 0.650506854057312,
|
| 1991 |
+
"learning_rate": 3.5173528632094355e-06,
|
| 1992 |
+
"loss": 0.1271,
|
| 1993 |
+
"step": 137000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 6.718131626520741,
|
| 1997 |
+
"grad_norm": 54.199398040771484,
|
| 1998 |
+
"learning_rate": 3.491363851363644e-06,
|
| 1999 |
+
"loss": 0.1315,
|
| 2000 |
+
"step": 137500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 6.742561196071725,
|
| 2004 |
+
"grad_norm": 23.973060607910156,
|
| 2005 |
+
"learning_rate": 3.465374839517852e-06,
|
| 2006 |
+
"loss": 0.1241,
|
| 2007 |
+
"step": 138000
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 6.76699076562271,
|
| 2011 |
+
"grad_norm": 0.1303454339504242,
|
| 2012 |
+
"learning_rate": 3.4393858276720605e-06,
|
| 2013 |
+
"loss": 0.1214,
|
| 2014 |
+
"step": 138500
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 6.791420335173695,
|
| 2018 |
+
"grad_norm": 0.046723511070013046,
|
| 2019 |
+
"learning_rate": 3.4133968158262686e-06,
|
| 2020 |
+
"loss": 0.1143,
|
| 2021 |
+
"step": 139000
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 6.815849904724679,
|
| 2025 |
+
"grad_norm": 6.235154628753662,
|
| 2026 |
+
"learning_rate": 3.3874078039804775e-06,
|
| 2027 |
+
"loss": 0.124,
|
| 2028 |
+
"step": 139500
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 6.840279474275663,
|
| 2032 |
+
"grad_norm": 0.06593639403581619,
|
| 2033 |
+
"learning_rate": 3.361418792134686e-06,
|
| 2034 |
+
"loss": 0.122,
|
| 2035 |
+
"step": 140000
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 6.864709043826648,
|
| 2039 |
+
"grad_norm": 0.6418370604515076,
|
| 2040 |
+
"learning_rate": 3.335429780288894e-06,
|
| 2041 |
+
"loss": 0.1232,
|
| 2042 |
+
"step": 140500
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 6.889138613377632,
|
| 2046 |
+
"grad_norm": 0.2320944368839264,
|
| 2047 |
+
"learning_rate": 3.3094407684431025e-06,
|
| 2048 |
+
"loss": 0.1232,
|
| 2049 |
+
"step": 141000
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 6.913568182928617,
|
| 2053 |
+
"grad_norm": 30.253524780273438,
|
| 2054 |
+
"learning_rate": 3.283451756597311e-06,
|
| 2055 |
+
"loss": 0.1298,
|
| 2056 |
+
"step": 141500
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 6.9379977524796015,
|
| 2060 |
+
"grad_norm": 261.870849609375,
|
| 2061 |
+
"learning_rate": 3.2574627447515195e-06,
|
| 2062 |
+
"loss": 0.1192,
|
| 2063 |
+
"step": 142000
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 6.962427322030586,
|
| 2067 |
+
"grad_norm": 0.349097341299057,
|
| 2068 |
+
"learning_rate": 3.2314737329057276e-06,
|
| 2069 |
+
"loss": 0.1236,
|
| 2070 |
+
"step": 142500
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 6.986856891581571,
|
| 2074 |
+
"grad_norm": 0.3783528208732605,
|
| 2075 |
+
"learning_rate": 3.205484721059936e-06,
|
| 2076 |
+
"loss": 0.1148,
|
| 2077 |
+
"step": 143000
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 7.0,
|
| 2081 |
+
"eval_accuracy": 0.9117218634052494,
|
| 2082 |
+
"eval_averaged_scores": 0.8971449328725574,
|
| 2083 |
+
"eval_f1": 0.8825680023398654,
|
| 2084 |
+
"eval_loss": 0.467751681804657,
|
| 2085 |
+
"eval_runtime": 20.1427,
|
| 2086 |
+
"eval_samples_per_second": 1806.36,
|
| 2087 |
+
"eval_steps_per_second": 112.944,
|
| 2088 |
+
"step": 143269
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 7.0112864611325545,
|
| 2092 |
+
"grad_norm": 1.4202431440353394,
|
| 2093 |
+
"learning_rate": 3.179495709214144e-06,
|
| 2094 |
+
"loss": 0.1088,
|
| 2095 |
+
"step": 143500
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 7.035716030683539,
|
| 2099 |
+
"grad_norm": 0.1592591255903244,
|
| 2100 |
+
"learning_rate": 3.153506697368353e-06,
|
| 2101 |
+
"loss": 0.0945,
|
| 2102 |
+
"step": 144000
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 7.060145600234524,
|
| 2106 |
+
"grad_norm": 8.698958396911621,
|
| 2107 |
+
"learning_rate": 3.1275176855225615e-06,
|
| 2108 |
+
"loss": 0.0984,
|
| 2109 |
+
"step": 144500
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 7.084575169785508,
|
| 2113 |
+
"grad_norm": 0.0554330088198185,
|
| 2114 |
+
"learning_rate": 3.1015286736767696e-06,
|
| 2115 |
+
"loss": 0.0925,
|
| 2116 |
+
"step": 145000
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 7.109004739336493,
|
| 2120 |
+
"grad_norm": 0.3783748745918274,
|
| 2121 |
+
"learning_rate": 3.075539661830978e-06,
|
| 2122 |
+
"loss": 0.0914,
|
| 2123 |
+
"step": 145500
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 7.133434308887478,
|
| 2127 |
+
"grad_norm": 0.33608478307724,
|
| 2128 |
+
"learning_rate": 3.049550649985186e-06,
|
| 2129 |
+
"loss": 0.0978,
|
| 2130 |
+
"step": 146000
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 7.157863878438462,
|
| 2134 |
+
"grad_norm": 0.36914902925491333,
|
| 2135 |
+
"learning_rate": 3.023561638139395e-06,
|
| 2136 |
+
"loss": 0.0923,
|
| 2137 |
+
"step": 146500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 7.182293447989446,
|
| 2141 |
+
"grad_norm": 0.05221285670995712,
|
| 2142 |
+
"learning_rate": 2.9975726262936036e-06,
|
| 2143 |
+
"loss": 0.0939,
|
| 2144 |
+
"step": 147000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 7.206723017540431,
|
| 2148 |
+
"grad_norm": 0.09707313030958176,
|
| 2149 |
+
"learning_rate": 2.9715836144478116e-06,
|
| 2150 |
+
"loss": 0.0925,
|
| 2151 |
+
"step": 147500
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 7.231152587091415,
|
| 2155 |
+
"grad_norm": 0.21917091310024261,
|
| 2156 |
+
"learning_rate": 2.94559460260202e-06,
|
| 2157 |
+
"loss": 0.0947,
|
| 2158 |
+
"step": 148000
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 7.2555821566424,
|
| 2162 |
+
"grad_norm": 0.08681820333003998,
|
| 2163 |
+
"learning_rate": 2.9196055907562286e-06,
|
| 2164 |
+
"loss": 0.0965,
|
| 2165 |
+
"step": 148500
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 7.2800117261933845,
|
| 2169 |
+
"grad_norm": 0.27925539016723633,
|
| 2170 |
+
"learning_rate": 2.893616578910437e-06,
|
| 2171 |
+
"loss": 0.1044,
|
| 2172 |
+
"step": 149000
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 7.304441295744369,
|
| 2176 |
+
"grad_norm": 33.41240692138672,
|
| 2177 |
+
"learning_rate": 2.867627567064645e-06,
|
| 2178 |
+
"loss": 0.0968,
|
| 2179 |
+
"step": 149500
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 7.328870865295354,
|
| 2183 |
+
"grad_norm": 142.32989501953125,
|
| 2184 |
+
"learning_rate": 2.8416385552188536e-06,
|
| 2185 |
+
"loss": 0.0998,
|
| 2186 |
+
"step": 150000
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 7.3533004348463376,
|
| 2190 |
+
"grad_norm": 0.03225061669945717,
|
| 2191 |
+
"learning_rate": 2.8156495433730617e-06,
|
| 2192 |
+
"loss": 0.1013,
|
| 2193 |
+
"step": 150500
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 7.377730004397322,
|
| 2197 |
+
"grad_norm": 0.08403612673282623,
|
| 2198 |
+
"learning_rate": 2.7896605315272706e-06,
|
| 2199 |
+
"loss": 0.1009,
|
| 2200 |
+
"step": 151000
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 7.402159573948307,
|
| 2204 |
+
"grad_norm": 0.3820052742958069,
|
| 2205 |
+
"learning_rate": 2.763671519681479e-06,
|
| 2206 |
+
"loss": 0.0963,
|
| 2207 |
+
"step": 151500
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 7.4265891434992914,
|
| 2211 |
+
"grad_norm": 16.261816024780273,
|
| 2212 |
+
"learning_rate": 2.737682507835687e-06,
|
| 2213 |
+
"loss": 0.1029,
|
| 2214 |
+
"step": 152000
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 7.451018713050276,
|
| 2218 |
+
"grad_norm": 0.331718385219574,
|
| 2219 |
+
"learning_rate": 2.7116934959898957e-06,
|
| 2220 |
+
"loss": 0.1041,
|
| 2221 |
+
"step": 152500
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 7.475448282601261,
|
| 2225 |
+
"grad_norm": 1.3978043794631958,
|
| 2226 |
+
"learning_rate": 2.685704484144104e-06,
|
| 2227 |
+
"loss": 0.0921,
|
| 2228 |
+
"step": 153000
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 7.499877852152245,
|
| 2232 |
+
"grad_norm": 0.04950063303112984,
|
| 2233 |
+
"learning_rate": 2.6597154722983127e-06,
|
| 2234 |
+
"loss": 0.0959,
|
| 2235 |
+
"step": 153500
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 7.52430742170323,
|
| 2239 |
+
"grad_norm": 0.14950354397296906,
|
| 2240 |
+
"learning_rate": 2.6337264604525207e-06,
|
| 2241 |
+
"loss": 0.0886,
|
| 2242 |
+
"step": 154000
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 7.548736991254214,
|
| 2246 |
+
"grad_norm": 37.28718566894531,
|
| 2247 |
+
"learning_rate": 2.607737448606729e-06,
|
| 2248 |
+
"loss": 0.1035,
|
| 2249 |
+
"step": 154500
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 7.573166560805198,
|
| 2253 |
+
"grad_norm": 0.17789612710475922,
|
| 2254 |
+
"learning_rate": 2.5817484367609373e-06,
|
| 2255 |
+
"loss": 0.0945,
|
| 2256 |
+
"step": 155000
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 7.597596130356183,
|
| 2260 |
+
"grad_norm": 0.2735101580619812,
|
| 2261 |
+
"learning_rate": 2.555759424915146e-06,
|
| 2262 |
+
"loss": 0.0996,
|
| 2263 |
+
"step": 155500
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 7.622025699907168,
|
| 2267 |
+
"grad_norm": 162.47628784179688,
|
| 2268 |
+
"learning_rate": 2.5297704130693547e-06,
|
| 2269 |
+
"loss": 0.1001,
|
| 2270 |
+
"step": 156000
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 7.646455269458152,
|
| 2274 |
+
"grad_norm": 0.21928128600120544,
|
| 2275 |
+
"learning_rate": 2.5037814012235627e-06,
|
| 2276 |
+
"loss": 0.1058,
|
| 2277 |
+
"step": 156500
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 7.670884839009137,
|
| 2281 |
+
"grad_norm": 8.425532341003418,
|
| 2282 |
+
"learning_rate": 2.4777923893777712e-06,
|
| 2283 |
+
"loss": 0.0884,
|
| 2284 |
+
"step": 157000
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 7.6953144085601215,
|
| 2288 |
+
"grad_norm": 0.1196184754371643,
|
| 2289 |
+
"learning_rate": 2.4518033775319797e-06,
|
| 2290 |
+
"loss": 0.1073,
|
| 2291 |
+
"step": 157500
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 7.719743978111106,
|
| 2295 |
+
"grad_norm": 18.718048095703125,
|
| 2296 |
+
"learning_rate": 2.425814365686188e-06,
|
| 2297 |
+
"loss": 0.0884,
|
| 2298 |
+
"step": 158000
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 7.74417354766209,
|
| 2302 |
+
"grad_norm": 37.17155075073242,
|
| 2303 |
+
"learning_rate": 2.3998253538403967e-06,
|
| 2304 |
+
"loss": 0.0945,
|
| 2305 |
+
"step": 158500
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 7.7686031172130745,
|
| 2309 |
+
"grad_norm": 0.38690435886383057,
|
| 2310 |
+
"learning_rate": 2.3738363419946048e-06,
|
| 2311 |
+
"loss": 0.0976,
|
| 2312 |
+
"step": 159000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 7.793032686764059,
|
| 2316 |
+
"grad_norm": 13.13317584991455,
|
| 2317 |
+
"learning_rate": 2.3478473301488133e-06,
|
| 2318 |
+
"loss": 0.0986,
|
| 2319 |
+
"step": 159500
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 7.817462256315044,
|
| 2323 |
+
"grad_norm": 0.05163729190826416,
|
| 2324 |
+
"learning_rate": 2.3218583183030217e-06,
|
| 2325 |
+
"loss": 0.1,
|
| 2326 |
+
"step": 160000
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 7.841891825866028,
|
| 2330 |
+
"grad_norm": 25.491252899169922,
|
| 2331 |
+
"learning_rate": 2.2958693064572302e-06,
|
| 2332 |
+
"loss": 0.0874,
|
| 2333 |
+
"step": 160500
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 7.866321395417013,
|
| 2337 |
+
"grad_norm": 0.3773481547832489,
|
| 2338 |
+
"learning_rate": 2.2698802946114383e-06,
|
| 2339 |
+
"loss": 0.1027,
|
| 2340 |
+
"step": 161000
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 7.890750964967998,
|
| 2344 |
+
"grad_norm": 13.568100929260254,
|
| 2345 |
+
"learning_rate": 2.243891282765647e-06,
|
| 2346 |
+
"loss": 0.1062,
|
| 2347 |
+
"step": 161500
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 7.915180534518981,
|
| 2351 |
+
"grad_norm": 24.115978240966797,
|
| 2352 |
+
"learning_rate": 2.2179022709198553e-06,
|
| 2353 |
+
"loss": 0.0934,
|
| 2354 |
+
"step": 162000
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 7.939610104069966,
|
| 2358 |
+
"grad_norm": 0.012055924162268639,
|
| 2359 |
+
"learning_rate": 2.1919132590740634e-06,
|
| 2360 |
+
"loss": 0.0854,
|
| 2361 |
+
"step": 162500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 7.964039673620951,
|
| 2365 |
+
"grad_norm": 0.3433281481266022,
|
| 2366 |
+
"learning_rate": 2.1659242472282723e-06,
|
| 2367 |
+
"loss": 0.1038,
|
| 2368 |
+
"step": 163000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 7.988469243171935,
|
| 2372 |
+
"grad_norm": 0.2032276839017868,
|
| 2373 |
+
"learning_rate": 2.1399352353824803e-06,
|
| 2374 |
+
"loss": 0.0992,
|
| 2375 |
+
"step": 163500
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 8.0,
|
| 2379 |
+
"eval_accuracy": 0.9103476707434383,
|
| 2380 |
+
"eval_averaged_scores": 0.8959594874193519,
|
| 2381 |
+
"eval_f1": 0.8815713040952657,
|
| 2382 |
+
"eval_loss": 0.5451585650444031,
|
| 2383 |
+
"eval_runtime": 20.179,
|
| 2384 |
+
"eval_samples_per_second": 1803.108,
|
| 2385 |
+
"eval_steps_per_second": 112.741,
|
| 2386 |
+
"step": 163736
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 8.01289881272292,
|
| 2390 |
+
"grad_norm": 0.012172297574579716,
|
| 2391 |
+
"learning_rate": 2.113946223536689e-06,
|
| 2392 |
+
"loss": 0.0873,
|
| 2393 |
+
"step": 164000
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 8.037328382273904,
|
| 2397 |
+
"grad_norm": 0.4155900180339813,
|
| 2398 |
+
"learning_rate": 2.0879572116908973e-06,
|
| 2399 |
+
"loss": 0.0777,
|
| 2400 |
+
"step": 164500
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 8.06175795182489,
|
| 2404 |
+
"grad_norm": 0.19762279093265533,
|
| 2405 |
+
"learning_rate": 2.061968199845106e-06,
|
| 2406 |
+
"loss": 0.0829,
|
| 2407 |
+
"step": 165000
|
| 2408 |
+
},
|
| 2409 |
+
{
|
| 2410 |
+
"epoch": 8.086187521375873,
|
| 2411 |
+
"grad_norm": 0.1475783735513687,
|
| 2412 |
+
"learning_rate": 2.035979187999314e-06,
|
| 2413 |
+
"loss": 0.0707,
|
| 2414 |
+
"step": 165500
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 8.110617090926858,
|
| 2418 |
+
"grad_norm": 145.56173706054688,
|
| 2419 |
+
"learning_rate": 2.0099901761535224e-06,
|
| 2420 |
+
"loss": 0.0751,
|
| 2421 |
+
"step": 166000
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"epoch": 8.135046660477842,
|
| 2425 |
+
"grad_norm": 0.041457872837781906,
|
| 2426 |
+
"learning_rate": 1.984001164307731e-06,
|
| 2427 |
+
"loss": 0.0671,
|
| 2428 |
+
"step": 166500
|
| 2429 |
+
},
|
| 2430 |
+
{
|
| 2431 |
+
"epoch": 8.159476230028828,
|
| 2432 |
+
"grad_norm": 5.121572971343994,
|
| 2433 |
+
"learning_rate": 1.9580121524619393e-06,
|
| 2434 |
+
"loss": 0.078,
|
| 2435 |
+
"step": 167000
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 8.183905799579811,
|
| 2439 |
+
"grad_norm": 0.2376221865415573,
|
| 2440 |
+
"learning_rate": 1.932023140616148e-06,
|
| 2441 |
+
"loss": 0.0842,
|
| 2442 |
+
"step": 167500
|
| 2443 |
+
},
|
| 2444 |
+
{
|
| 2445 |
+
"epoch": 8.208335369130795,
|
| 2446 |
+
"grad_norm": 27.775150299072266,
|
| 2447 |
+
"learning_rate": 1.9060341287703559e-06,
|
| 2448 |
+
"loss": 0.0767,
|
| 2449 |
+
"step": 168000
|
| 2450 |
+
},
|
| 2451 |
+
{
|
| 2452 |
+
"epoch": 8.23276493868178,
|
| 2453 |
+
"grad_norm": 0.18645720183849335,
|
| 2454 |
+
"learning_rate": 1.8800451169245646e-06,
|
| 2455 |
+
"loss": 0.0802,
|
| 2456 |
+
"step": 168500
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 8.257194508232764,
|
| 2460 |
+
"grad_norm": 7.003711700439453,
|
| 2461 |
+
"learning_rate": 1.8540561050787729e-06,
|
| 2462 |
+
"loss": 0.0803,
|
| 2463 |
+
"step": 169000
|
| 2464 |
+
},
|
| 2465 |
+
{
|
| 2466 |
+
"epoch": 8.28162407778375,
|
| 2467 |
+
"grad_norm": 44.9241828918457,
|
| 2468 |
+
"learning_rate": 1.8280670932329814e-06,
|
| 2469 |
+
"loss": 0.08,
|
| 2470 |
+
"step": 169500
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 8.306053647334734,
|
| 2474 |
+
"grad_norm": 0.3100529909133911,
|
| 2475 |
+
"learning_rate": 1.8020780813871896e-06,
|
| 2476 |
+
"loss": 0.0807,
|
| 2477 |
+
"step": 170000
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 8.33048321688572,
|
| 2481 |
+
"grad_norm": 1.1184957027435303,
|
| 2482 |
+
"learning_rate": 1.776089069541398e-06,
|
| 2483 |
+
"loss": 0.0744,
|
| 2484 |
+
"step": 170500
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 8.354912786436703,
|
| 2488 |
+
"grad_norm": 0.22304557263851166,
|
| 2489 |
+
"learning_rate": 1.7501000576956064e-06,
|
| 2490 |
+
"loss": 0.0784,
|
| 2491 |
+
"step": 171000
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 8.379342355987687,
|
| 2495 |
+
"grad_norm": 0.036572836339473724,
|
| 2496 |
+
"learning_rate": 1.7241110458498147e-06,
|
| 2497 |
+
"loss": 0.077,
|
| 2498 |
+
"step": 171500
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 8.403771925538672,
|
| 2502 |
+
"grad_norm": 0.1464405059814453,
|
| 2503 |
+
"learning_rate": 1.6981220340040234e-06,
|
| 2504 |
+
"loss": 0.0834,
|
| 2505 |
+
"step": 172000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 8.428201495089656,
|
| 2509 |
+
"grad_norm": 0.08662763237953186,
|
| 2510 |
+
"learning_rate": 1.6721330221582317e-06,
|
| 2511 |
+
"loss": 0.0787,
|
| 2512 |
+
"step": 172500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 8.452631064640642,
|
| 2516 |
+
"grad_norm": 1.44623601436615,
|
| 2517 |
+
"learning_rate": 1.6461440103124402e-06,
|
| 2518 |
+
"loss": 0.0726,
|
| 2519 |
+
"step": 173000
|
| 2520 |
+
},
|
| 2521 |
+
{
|
| 2522 |
+
"epoch": 8.477060634191625,
|
| 2523 |
+
"grad_norm": 14.024663925170898,
|
| 2524 |
+
"learning_rate": 1.6201549984666484e-06,
|
| 2525 |
+
"loss": 0.0907,
|
| 2526 |
+
"step": 173500
|
| 2527 |
+
},
|
| 2528 |
+
{
|
| 2529 |
+
"epoch": 8.50149020374261,
|
| 2530 |
+
"grad_norm": 0.02955411933362484,
|
| 2531 |
+
"learning_rate": 1.5941659866208567e-06,
|
| 2532 |
+
"loss": 0.0836,
|
| 2533 |
+
"step": 174000
|
| 2534 |
+
},
|
| 2535 |
+
{
|
| 2536 |
+
"epoch": 8.525919773293595,
|
| 2537 |
+
"grad_norm": 0.1095781922340393,
|
| 2538 |
+
"learning_rate": 1.5681769747750652e-06,
|
| 2539 |
+
"loss": 0.0881,
|
| 2540 |
+
"step": 174500
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 8.550349342844578,
|
| 2544 |
+
"grad_norm": 0.8175545930862427,
|
| 2545 |
+
"learning_rate": 1.5421879629292735e-06,
|
| 2546 |
+
"loss": 0.0858,
|
| 2547 |
+
"step": 175000
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 8.574778912395564,
|
| 2551 |
+
"grad_norm": 0.0535324402153492,
|
| 2552 |
+
"learning_rate": 1.5161989510834822e-06,
|
| 2553 |
+
"loss": 0.0781,
|
| 2554 |
+
"step": 175500
|
| 2555 |
+
},
|
| 2556 |
+
{
|
| 2557 |
+
"epoch": 8.599208481946548,
|
| 2558 |
+
"grad_norm": 0.0196294616907835,
|
| 2559 |
+
"learning_rate": 1.4902099392376905e-06,
|
| 2560 |
+
"loss": 0.0991,
|
| 2561 |
+
"step": 176000
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 8.623638051497533,
|
| 2565 |
+
"grad_norm": 3.7140159606933594,
|
| 2566 |
+
"learning_rate": 1.464220927391899e-06,
|
| 2567 |
+
"loss": 0.0747,
|
| 2568 |
+
"step": 176500
|
| 2569 |
+
},
|
| 2570 |
+
{
|
| 2571 |
+
"epoch": 8.648067621048517,
|
| 2572 |
+
"grad_norm": 101.7074966430664,
|
| 2573 |
+
"learning_rate": 1.4382319155461072e-06,
|
| 2574 |
+
"loss": 0.0914,
|
| 2575 |
+
"step": 177000
|
| 2576 |
+
},
|
| 2577 |
+
{
|
| 2578 |
+
"epoch": 8.672497190599502,
|
| 2579 |
+
"grad_norm": 0.013971562497317791,
|
| 2580 |
+
"learning_rate": 1.4122429037003157e-06,
|
| 2581 |
+
"loss": 0.0681,
|
| 2582 |
+
"step": 177500
|
| 2583 |
+
},
|
| 2584 |
+
{
|
| 2585 |
+
"epoch": 8.696926760150486,
|
| 2586 |
+
"grad_norm": 164.79086303710938,
|
| 2587 |
+
"learning_rate": 1.386253891854524e-06,
|
| 2588 |
+
"loss": 0.0789,
|
| 2589 |
+
"step": 178000
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 8.721356329701472,
|
| 2593 |
+
"grad_norm": 0.1486426293849945,
|
| 2594 |
+
"learning_rate": 1.3602648800087323e-06,
|
| 2595 |
+
"loss": 0.0847,
|
| 2596 |
+
"step": 178500
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 8.745785899252455,
|
| 2600 |
+
"grad_norm": 32.686614990234375,
|
| 2601 |
+
"learning_rate": 1.3342758681629408e-06,
|
| 2602 |
+
"loss": 0.0914,
|
| 2603 |
+
"step": 179000
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 8.770215468803439,
|
| 2607 |
+
"grad_norm": 0.8406556844711304,
|
| 2608 |
+
"learning_rate": 1.308286856317149e-06,
|
| 2609 |
+
"loss": 0.0742,
|
| 2610 |
+
"step": 179500
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 8.794645038354425,
|
| 2614 |
+
"grad_norm": 0.07689370214939117,
|
| 2615 |
+
"learning_rate": 1.2822978444713577e-06,
|
| 2616 |
+
"loss": 0.0791,
|
| 2617 |
+
"step": 180000
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 8.819074607905408,
|
| 2621 |
+
"grad_norm": 17.384950637817383,
|
| 2622 |
+
"learning_rate": 1.256308832625566e-06,
|
| 2623 |
+
"loss": 0.0764,
|
| 2624 |
+
"step": 180500
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 8.843504177456394,
|
| 2628 |
+
"grad_norm": 29.185670852661133,
|
| 2629 |
+
"learning_rate": 1.2303198207797743e-06,
|
| 2630 |
+
"loss": 0.0862,
|
| 2631 |
+
"step": 181000
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"epoch": 8.867933747007378,
|
| 2635 |
+
"grad_norm": 58.11179733276367,
|
| 2636 |
+
"learning_rate": 1.2043308089339828e-06,
|
| 2637 |
+
"loss": 0.0862,
|
| 2638 |
+
"step": 181500
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 8.892363316558363,
|
| 2642 |
+
"grad_norm": 0.12057217955589294,
|
| 2643 |
+
"learning_rate": 1.1783417970881913e-06,
|
| 2644 |
+
"loss": 0.08,
|
| 2645 |
+
"step": 182000
|
| 2646 |
+
},
|
| 2647 |
+
{
|
| 2648 |
+
"epoch": 8.916792886109347,
|
| 2649 |
+
"grad_norm": 67.25086212158203,
|
| 2650 |
+
"learning_rate": 1.1523527852423996e-06,
|
| 2651 |
+
"loss": 0.0881,
|
| 2652 |
+
"step": 182500
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 8.94122245566033,
|
| 2656 |
+
"grad_norm": 0.04171622171998024,
|
| 2657 |
+
"learning_rate": 1.126363773396608e-06,
|
| 2658 |
+
"loss": 0.0831,
|
| 2659 |
+
"step": 183000
|
| 2660 |
+
},
|
| 2661 |
+
{
|
| 2662 |
+
"epoch": 8.965652025211316,
|
| 2663 |
+
"grad_norm": 0.3835027515888214,
|
| 2664 |
+
"learning_rate": 1.1003747615508165e-06,
|
| 2665 |
+
"loss": 0.0768,
|
| 2666 |
+
"step": 183500
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 8.9900815947623,
|
| 2670 |
+
"grad_norm": 0.16164089739322662,
|
| 2671 |
+
"learning_rate": 1.0743857497050248e-06,
|
| 2672 |
+
"loss": 0.0847,
|
| 2673 |
+
"step": 184000
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 9.0,
|
| 2677 |
+
"eval_accuracy": 0.9122440566167377,
|
| 2678 |
+
"eval_averaged_scores": 0.8971997179189981,
|
| 2679 |
+
"eval_f1": 0.8821553792212585,
|
| 2680 |
+
"eval_loss": 0.5808023810386658,
|
| 2681 |
+
"eval_runtime": 20.2131,
|
| 2682 |
+
"eval_samples_per_second": 1800.067,
|
| 2683 |
+
"eval_steps_per_second": 112.551,
|
| 2684 |
+
"step": 184203
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 9.014511164313285,
|
| 2688 |
+
"grad_norm": 0.031089797616004944,
|
| 2689 |
+
"learning_rate": 1.048396737859233e-06,
|
| 2690 |
+
"loss": 0.072,
|
| 2691 |
+
"step": 184500
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 9.038940733864269,
|
| 2695 |
+
"grad_norm": 0.0398230142891407,
|
| 2696 |
+
"learning_rate": 1.0224077260134416e-06,
|
| 2697 |
+
"loss": 0.0608,
|
| 2698 |
+
"step": 185000
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 9.063370303415255,
|
| 2702 |
+
"grad_norm": 0.15601158142089844,
|
| 2703 |
+
"learning_rate": 9.9641871416765e-07,
|
| 2704 |
+
"loss": 0.0689,
|
| 2705 |
+
"step": 185500
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 9.087799872966238,
|
| 2709 |
+
"grad_norm": 1.7415249347686768,
|
| 2710 |
+
"learning_rate": 9.704297023218584e-07,
|
| 2711 |
+
"loss": 0.0676,
|
| 2712 |
+
"step": 186000
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 9.112229442517222,
|
| 2716 |
+
"grad_norm": 0.036363422870635986,
|
| 2717 |
+
"learning_rate": 9.444406904760668e-07,
|
| 2718 |
+
"loss": 0.0696,
|
| 2719 |
+
"step": 186500
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 9.136659012068208,
|
| 2723 |
+
"grad_norm": 16.439847946166992,
|
| 2724 |
+
"learning_rate": 9.184516786302752e-07,
|
| 2725 |
+
"loss": 0.0704,
|
| 2726 |
+
"step": 187000
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 9.161088581619191,
|
| 2730 |
+
"grad_norm": 0.16665898263454437,
|
| 2731 |
+
"learning_rate": 8.924626667844836e-07,
|
| 2732 |
+
"loss": 0.0706,
|
| 2733 |
+
"step": 187500
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 9.185518151170177,
|
| 2737 |
+
"grad_norm": 0.006615887396037579,
|
| 2738 |
+
"learning_rate": 8.664736549386919e-07,
|
| 2739 |
+
"loss": 0.0718,
|
| 2740 |
+
"step": 188000
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 9.20994772072116,
|
| 2744 |
+
"grad_norm": 0.015520376153290272,
|
| 2745 |
+
"learning_rate": 8.404846430929004e-07,
|
| 2746 |
+
"loss": 0.0639,
|
| 2747 |
+
"step": 188500
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 9.234377290272146,
|
| 2751 |
+
"grad_norm": 0.18699295818805695,
|
| 2752 |
+
"learning_rate": 8.144956312471088e-07,
|
| 2753 |
+
"loss": 0.0671,
|
| 2754 |
+
"step": 189000
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 9.25880685982313,
|
| 2758 |
+
"grad_norm": 208.81222534179688,
|
| 2759 |
+
"learning_rate": 7.885066194013171e-07,
|
| 2760 |
+
"loss": 0.0577,
|
| 2761 |
+
"step": 189500
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 9.283236429374114,
|
| 2765 |
+
"grad_norm": 35.29654312133789,
|
| 2766 |
+
"learning_rate": 7.625176075555256e-07,
|
| 2767 |
+
"loss": 0.0754,
|
| 2768 |
+
"step": 190000
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 9.3076659989251,
|
| 2772 |
+
"grad_norm": 0.024813145399093628,
|
| 2773 |
+
"learning_rate": 7.36528595709734e-07,
|
| 2774 |
+
"loss": 0.073,
|
| 2775 |
+
"step": 190500
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 9.332095568476083,
|
| 2779 |
+
"grad_norm": 0.0019479466136544943,
|
| 2780 |
+
"learning_rate": 7.105395838639424e-07,
|
| 2781 |
+
"loss": 0.0621,
|
| 2782 |
+
"step": 191000
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 9.356525138027068,
|
| 2786 |
+
"grad_norm": 0.034291163086891174,
|
| 2787 |
+
"learning_rate": 6.845505720181508e-07,
|
| 2788 |
+
"loss": 0.0646,
|
| 2789 |
+
"step": 191500
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 9.380954707578052,
|
| 2793 |
+
"grad_norm": 21.998003005981445,
|
| 2794 |
+
"learning_rate": 6.585615601723591e-07,
|
| 2795 |
+
"loss": 0.058,
|
| 2796 |
+
"step": 192000
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 9.405384277129038,
|
| 2800 |
+
"grad_norm": 0.04920468479394913,
|
| 2801 |
+
"learning_rate": 6.325725483265676e-07,
|
| 2802 |
+
"loss": 0.0834,
|
| 2803 |
+
"step": 192500
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 9.429813846680021,
|
| 2807 |
+
"grad_norm": 0.0892450362443924,
|
| 2808 |
+
"learning_rate": 6.065835364807759e-07,
|
| 2809 |
+
"loss": 0.0666,
|
| 2810 |
+
"step": 193000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 9.454243416231005,
|
| 2814 |
+
"grad_norm": 0.2844100892543793,
|
| 2815 |
+
"learning_rate": 5.805945246349843e-07,
|
| 2816 |
+
"loss": 0.0583,
|
| 2817 |
+
"step": 193500
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 9.47867298578199,
|
| 2821 |
+
"grad_norm": 0.26199427247047424,
|
| 2822 |
+
"learning_rate": 5.546055127891928e-07,
|
| 2823 |
+
"loss": 0.0719,
|
| 2824 |
+
"step": 194000
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 9.503102555332974,
|
| 2828 |
+
"grad_norm": 0.016245270147919655,
|
| 2829 |
+
"learning_rate": 5.286165009434012e-07,
|
| 2830 |
+
"loss": 0.072,
|
| 2831 |
+
"step": 194500
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 9.52753212488396,
|
| 2835 |
+
"grad_norm": 0.2348903864622116,
|
| 2836 |
+
"learning_rate": 5.026274890976096e-07,
|
| 2837 |
+
"loss": 0.0594,
|
| 2838 |
+
"step": 195000
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 9.551961694434944,
|
| 2842 |
+
"grad_norm": 0.00560354720801115,
|
| 2843 |
+
"learning_rate": 4.7663847725181796e-07,
|
| 2844 |
+
"loss": 0.0601,
|
| 2845 |
+
"step": 195500
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 9.57639126398593,
|
| 2849 |
+
"grad_norm": 0.20233391225337982,
|
| 2850 |
+
"learning_rate": 4.5064946540602635e-07,
|
| 2851 |
+
"loss": 0.0597,
|
| 2852 |
+
"step": 196000
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 9.600820833536913,
|
| 2856 |
+
"grad_norm": 0.028764836490154266,
|
| 2857 |
+
"learning_rate": 4.246604535602348e-07,
|
| 2858 |
+
"loss": 0.0748,
|
| 2859 |
+
"step": 196500
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 9.625250403087897,
|
| 2863 |
+
"grad_norm": 0.03489026054739952,
|
| 2864 |
+
"learning_rate": 3.986714417144431e-07,
|
| 2865 |
+
"loss": 0.0593,
|
| 2866 |
+
"step": 197000
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 9.649679972638882,
|
| 2870 |
+
"grad_norm": 0.17677338421344757,
|
| 2871 |
+
"learning_rate": 3.7268242986865155e-07,
|
| 2872 |
+
"loss": 0.0575,
|
| 2873 |
+
"step": 197500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 9.674109542189866,
|
| 2877 |
+
"grad_norm": 0.006993240211158991,
|
| 2878 |
+
"learning_rate": 3.4669341802286e-07,
|
| 2879 |
+
"loss": 0.059,
|
| 2880 |
+
"step": 198000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 9.698539111740851,
|
| 2884 |
+
"grad_norm": 0.5008031129837036,
|
| 2885 |
+
"learning_rate": 3.207044061770684e-07,
|
| 2886 |
+
"loss": 0.0728,
|
| 2887 |
+
"step": 198500
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 9.722968681291835,
|
| 2891 |
+
"grad_norm": 0.20838424563407898,
|
| 2892 |
+
"learning_rate": 2.9471539433127676e-07,
|
| 2893 |
+
"loss": 0.0706,
|
| 2894 |
+
"step": 199000
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 9.74739825084282,
|
| 2898 |
+
"grad_norm": 0.023256618529558182,
|
| 2899 |
+
"learning_rate": 2.6872638248548514e-07,
|
| 2900 |
+
"loss": 0.0704,
|
| 2901 |
+
"step": 199500
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 9.771827820393804,
|
| 2905 |
+
"grad_norm": 0.2845751643180847,
|
| 2906 |
+
"learning_rate": 2.427373706396936e-07,
|
| 2907 |
+
"loss": 0.0665,
|
| 2908 |
+
"step": 200000
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 9.796257389944788,
|
| 2912 |
+
"grad_norm": 0.04077767953276634,
|
| 2913 |
+
"learning_rate": 2.1674835879390194e-07,
|
| 2914 |
+
"loss": 0.0609,
|
| 2915 |
+
"step": 200500
|
| 2916 |
+
},
|
| 2917 |
+
{
|
| 2918 |
+
"epoch": 9.820686959495774,
|
| 2919 |
+
"grad_norm": 0.01134885661303997,
|
| 2920 |
+
"learning_rate": 1.9075934694811035e-07,
|
| 2921 |
+
"loss": 0.0593,
|
| 2922 |
+
"step": 201000
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 9.845116529046757,
|
| 2926 |
+
"grad_norm": 0.018247269093990326,
|
| 2927 |
+
"learning_rate": 1.6477033510231873e-07,
|
| 2928 |
+
"loss": 0.0657,
|
| 2929 |
+
"step": 201500
|
| 2930 |
+
},
|
| 2931 |
+
{
|
| 2932 |
+
"epoch": 9.869546098597743,
|
| 2933 |
+
"grad_norm": 0.008613000623881817,
|
| 2934 |
+
"learning_rate": 1.3878132325652717e-07,
|
| 2935 |
+
"loss": 0.0563,
|
| 2936 |
+
"step": 202000
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 9.893975668148727,
|
| 2940 |
+
"grad_norm": 0.2736217975616455,
|
| 2941 |
+
"learning_rate": 1.1279231141073555e-07,
|
| 2942 |
+
"loss": 0.0709,
|
| 2943 |
+
"step": 202500
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 9.918405237699712,
|
| 2947 |
+
"grad_norm": 0.02775813639163971,
|
| 2948 |
+
"learning_rate": 8.680329956494395e-08,
|
| 2949 |
+
"loss": 0.0677,
|
| 2950 |
+
"step": 203000
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 9.942834807250696,
|
| 2954 |
+
"grad_norm": 0.06576963514089584,
|
| 2955 |
+
"learning_rate": 6.081428771915235e-08,
|
| 2956 |
+
"loss": 0.0685,
|
| 2957 |
+
"step": 203500
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 9.967264376801682,
|
| 2961 |
+
"grad_norm": 0.1034393236041069,
|
| 2962 |
+
"learning_rate": 3.4825275873360744e-08,
|
| 2963 |
+
"loss": 0.0677,
|
| 2964 |
+
"step": 204000
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 9.991693946352665,
|
| 2968 |
+
"grad_norm": 0.14732515811920166,
|
| 2969 |
+
"learning_rate": 8.836264027569144e-09,
|
| 2970 |
+
"loss": 0.0649,
|
| 2971 |
+
"step": 204500
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 10.0,
|
| 2975 |
+
"eval_accuracy": 0.9122165727635014,
|
| 2976 |
+
"eval_averaged_scores": 0.8973993679198728,
|
| 2977 |
+
"eval_f1": 0.8825821630762444,
|
| 2978 |
+
"eval_loss": 0.6054433584213257,
|
| 2979 |
+
"eval_runtime": 20.1824,
|
| 2980 |
+
"eval_samples_per_second": 1802.808,
|
| 2981 |
+
"eval_steps_per_second": 112.722,
|
| 2982 |
+
"step": 204670
|
| 2983 |
+
},
|
| 2984 |
+
{
|
| 2985 |
+
"epoch": 10.0,
|
| 2986 |
+
"step": 204670,
|
| 2987 |
+
"total_flos": 1.0020319530941755e+17,
|
| 2988 |
+
"train_loss": 0.16519966338430658,
|
| 2989 |
+
"train_runtime": 10940.7303,
|
| 2990 |
+
"train_samples_per_second": 299.305,
|
| 2991 |
+
"train_steps_per_second": 18.707
|
| 2992 |
+
}
|
| 2993 |
+
],
|
| 2994 |
+
"logging_steps": 500,
|
| 2995 |
+
"max_steps": 204670,
|
| 2996 |
+
"num_input_tokens_seen": 0,
|
| 2997 |
+
"num_train_epochs": 10,
|
| 2998 |
+
"save_steps": 500,
|
| 2999 |
+
"stateful_callbacks": {
|
| 3000 |
+
"TrainerControl": {
|
| 3001 |
+
"args": {
|
| 3002 |
+
"should_epoch_stop": false,
|
| 3003 |
+
"should_evaluate": false,
|
| 3004 |
+
"should_log": false,
|
| 3005 |
+
"should_save": true,
|
| 3006 |
+
"should_training_stop": true
|
| 3007 |
+
},
|
| 3008 |
+
"attributes": {}
|
| 3009 |
+
}
|
| 3010 |
+
},
|
| 3011 |
+
"total_flos": 1.0020319530941755e+17,
|
| 3012 |
+
"train_batch_size": 16,
|
| 3013 |
+
"trial_name": null,
|
| 3014 |
+
"trial_params": null
|
| 3015 |
+
}
|
qqp/roberta-base_lr1e-05/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d174ebffbf75c9a4cdd0fca75a1361a2b5e7f00a138154388f45bef5abbfe145
|
| 3 |
+
size 5240
|
qqp/roberta-base_lr1e-05/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/roberta-large_lr1e-05/classifier_head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d8912172603eda6ba4aab6217d6d3fa0a65f5706119244f8263aa081b15eb72
|
| 3 |
+
size 4210000
|
qqp/roberta-large_lr1e-05/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/projects/shikexuan/nlu_model/roberta-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.44.1",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
qqp/roberta-large_lr1e-05/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qqp/roberta-large_lr1e-05/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5f02ef21b52d00af34afdb2b98e08baf1caff1b1ebaad568bb2404ff861c24b
|
| 3 |
+
size 1421495416
|