Hugging Face's logo

Multi-Domain-Expert-Learning
/
expert-min-pile-instruct-0.1

Text Generation
Transformers
PyTorch
gpt_neox
Generated from Trainer
text-generation-inference
Model card Files
xet
Community
1
P1ayer-1 commited on
Commit
43d5cb9
·
1 Parent(s): 4cb32ba

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. .gitattributes +2 -34
  2. .gitignore +2 -0
  3. README.md +71 -0
  4. all_results.json +15 -0
  5. config.json +25 -0
  6. eval_results.json +10 -0
  7. generation_config.json +6 -0
  8. pytorch_model.bin +3 -0
  9. special_tokens_map.json +9 -0
  10. tokenizer.json +0 -0
  11. tokenizer_config.json +10 -0
  12. train_results.json +8 -0
  13. trainer_state.json +3895 -0
  14. training_args.bin +3 -0
.gitattributes CHANGED
@@ -1,34 +1,2 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tflite filter=lfs diff=lfs merge=lfs -text
29
- *.tgz filter=lfs diff=lfs merge=lfs -text
30
- *.wasm filter=lfs diff=lfs merge=lfs -text
31
- *.xz filter=lfs diff=lfs merge=lfs -text
32
- *.zip filter=lfs diff=lfs merge=lfs -text
33
- *.zst filter=lfs diff=lfs merge=lfs -text
34
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
1
+ pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
2
+ training_args.bin filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
.gitignore ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ checkpoint-*/
2
+ runs/
README.md ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ datasets:
5
+ - /pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pile-instruct
6
+ metrics:
7
+ - accuracy
8
+ model-index:
9
+ - name: layer_13,14,15
10
+ results:
11
+ - task:
12
+ name: Causal Language Modeling
13
+ type: text-generation
14
+ dataset:
15
+ name: /pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pile-instruct
16
+ type: /pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pile-instruct
17
+ metrics:
18
+ - name: Accuracy
19
+ type: accuracy
20
+ value: 1.0
21
+ ---
22
+
23
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
24
+ should probably proofread and complete it, then remove this comment. -->
25
+
26
+ # layer_13,14,15
27
+
28
+ This model is a fine-tuned version of [/pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pythia-deduped-1b-chat-base/](https://huggingface.co//pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pythia-deduped-1b-chat-base/) on the /pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pile-instruct dataset.
29
+ It achieves the following results on the evaluation set:
30
+ - Loss: 5.0430
31
+ - Accuracy: 1.0
32
+
33
+ ## Model description
34
+
35
+ More information needed
36
+
37
+ ## Intended uses & limitations
38
+
39
+ More information needed
40
+
41
+ ## Training and evaluation data
42
+
43
+ More information needed
44
+
45
+ ## Training procedure
46
+
47
+ ### Training hyperparameters
48
+
49
+ The following hyperparameters were used during training:
50
+ - learning_rate: 0.0001
51
+ - train_batch_size: 24
52
+ - eval_batch_size: 6
53
+ - seed: 42
54
+ - distributed_type: multi-GPU
55
+ - num_devices: 4
56
+ - total_train_batch_size: 96
57
+ - total_eval_batch_size: 24
58
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
+ - lr_scheduler_type: linear
60
+ - training_steps: 12000
61
+
62
+ ### Training results
63
+
64
+
65
+
66
+ ### Framework versions
67
+
68
+ - Transformers 4.27.0
69
+ - Pytorch 1.12.1+gitcb6c422
70
+ - Datasets 2.11.0
71
+ - Tokenizers 0.13.3
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.98,
3
+ "eval_accuracy": 1.0,
4
+ "eval_loss": 5.04296875,
5
+ "eval_runtime": 333.6725,
6
+ "eval_samples": 10329,
7
+ "eval_samples_per_second": 30.956,
8
+ "eval_steps_per_second": 1.292,
9
+ "perplexity": 154.92927926604594,
10
+ "train_loss": 0.2671610107421875,
11
+ "train_runtime": 4071.6945,
12
+ "train_samples": 192539,
13
+ "train_samples_per_second": 282.929,
14
+ "train_steps_per_second": 2.947
15
+ }
config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pythia-deduped-1b-chat-base/",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "bos_token_id": 0,
7
+ "eos_token_id": 0,
8
+ "hidden_act": "gelu",
9
+ "hidden_size": 2048,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 8192,
12
+ "layer_norm_eps": 1e-05,
13
+ "max_position_embeddings": 2048,
14
+ "model_type": "gpt_neox",
15
+ "num_attention_heads": 8,
16
+ "num_hidden_layers": 16,
17
+ "rotary_emb_base": 10000,
18
+ "rotary_pct": 0.25,
19
+ "tie_word_embeddings": false,
20
+ "torch_dtype": "float16",
21
+ "transformers_version": "4.27.0",
22
+ "use_cache": true,
23
+ "use_parallel_residual": true,
24
+ "vocab_size": 50279
25
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.98,
3
+ "eval_accuracy": 1.0,
4
+ "eval_loss": 5.04296875,
5
+ "eval_runtime": 333.6725,
6
+ "eval_samples": 10329,
7
+ "eval_samples_per_second": 30.956,
8
+ "eval_steps_per_second": 1.292,
9
+ "perplexity": 154.92927926604594
10
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "transformers_version": "4.27.0"
6
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f315973a9778dc6521c86233e697aeffeb50b4f1e6ebefac5e778b0b87800649
3
+ size 2090541917
special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<human>",
4
+ "<bot>"
5
+ ],
6
+ "bos_token": "<|endoftext|>",
7
+ "eos_token": "<|endoftext|>",
8
+ "unk_token": "<|endoftext|>"
9
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 1000000000000000019884624838656,
7
+ "special_tokens_map_file": "/pfs/lustrep4/scratch/project_462000259/noah/instruct_1bil/transfer/pythia-deduped-1b-chat-base/special_tokens_map.json",
8
+ "tokenizer_class": "GPTNeoXTokenizer",
9
+ "unk_token": "<|endoftext|>"
10
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.98,
3
+ "train_loss": 0.2671610107421875,
4
+ "train_runtime": 4071.6945,
5
+ "train_samples": 192539,
6
+ "train_samples_per_second": 282.929,
7
+ "train_steps_per_second": 2.947
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,3895 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.982053838484546,
5
+ "global_step": 12000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 9.987497916319387e-05,
13
+ "loss": 8.1937,
14
+ "step": 20
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 9.970828471411902e-05,
19
+ "loss": 5.8372,
20
+ "step": 40
21
+ },
22
+ {
23
+ "epoch": 0.03,
24
+ "learning_rate": 9.954159026504419e-05,
25
+ "loss": 5.7086,
26
+ "step": 60
27
+ },
28
+ {
29
+ "epoch": 0.04,
30
+ "learning_rate": 9.937489581596934e-05,
31
+ "loss": 5.5107,
32
+ "step": 80
33
+ },
34
+ {
35
+ "epoch": 0.05,
36
+ "learning_rate": 9.920820136689449e-05,
37
+ "loss": 5.3996,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.06,
42
+ "learning_rate": 9.904150691781964e-05,
43
+ "loss": 5.2727,
44
+ "step": 120
45
+ },
46
+ {
47
+ "epoch": 0.07,
48
+ "learning_rate": 9.887481246874479e-05,
49
+ "loss": 5.2045,
50
+ "step": 140
51
+ },
52
+ {
53
+ "epoch": 0.08,
54
+ "learning_rate": 9.870811801966994e-05,
55
+ "loss": 5.1715,
56
+ "step": 160
57
+ },
58
+ {
59
+ "epoch": 0.09,
60
+ "learning_rate": 9.85414235705951e-05,
61
+ "loss": 5.1009,
62
+ "step": 180
63
+ },
64
+ {
65
+ "epoch": 0.1,
66
+ "learning_rate": 9.837472912152026e-05,
67
+ "loss": 5.0439,
68
+ "step": 200
69
+ },
70
+ {
71
+ "epoch": 0.11,
72
+ "learning_rate": 9.820803467244541e-05,
73
+ "loss": 5.0172,
74
+ "step": 220
75
+ },
76
+ {
77
+ "epoch": 0.12,
78
+ "learning_rate": 9.804134022337057e-05,
79
+ "loss": 4.9717,
80
+ "step": 240
81
+ },
82
+ {
83
+ "epoch": 0.13,
84
+ "learning_rate": 9.787464577429572e-05,
85
+ "loss": 4.936,
86
+ "step": 260
87
+ },
88
+ {
89
+ "epoch": 0.14,
90
+ "learning_rate": 9.770795132522087e-05,
91
+ "loss": 4.8993,
92
+ "step": 280
93
+ },
94
+ {
95
+ "epoch": 0.15,
96
+ "learning_rate": 9.754125687614603e-05,
97
+ "loss": 4.8823,
98
+ "step": 300
99
+ },
100
+ {
101
+ "epoch": 0.16,
102
+ "learning_rate": 9.737456242707118e-05,
103
+ "loss": 4.8529,
104
+ "step": 320
105
+ },
106
+ {
107
+ "epoch": 0.17,
108
+ "learning_rate": 9.720786797799634e-05,
109
+ "loss": 4.8367,
110
+ "step": 340
111
+ },
112
+ {
113
+ "epoch": 0.18,
114
+ "learning_rate": 9.704117352892149e-05,
115
+ "loss": 4.8028,
116
+ "step": 360
117
+ },
118
+ {
119
+ "epoch": 0.19,
120
+ "learning_rate": 9.687447907984664e-05,
121
+ "loss": 4.7783,
122
+ "step": 380
123
+ },
124
+ {
125
+ "epoch": 0.2,
126
+ "learning_rate": 9.670778463077181e-05,
127
+ "loss": 4.7817,
128
+ "step": 400
129
+ },
130
+ {
131
+ "epoch": 0.2,
132
+ "eval_validation_accuracy": 0.6666666666666666,
133
+ "eval_validation_loss": 7.4375,
134
+ "eval_validation_runtime": 333.0271,
135
+ "eval_validation_samples_per_second": 31.015,
136
+ "eval_validation_steps_per_second": 1.294,
137
+ "step": 400
138
+ },
139
+ {
140
+ "epoch": 0.21,
141
+ "learning_rate": 9.654109018169696e-05,
142
+ "loss": 4.7515,
143
+ "step": 420
144
+ },
145
+ {
146
+ "epoch": 0.22,
147
+ "learning_rate": 9.637439573262211e-05,
148
+ "loss": 4.7273,
149
+ "step": 440
150
+ },
151
+ {
152
+ "epoch": 0.23,
153
+ "learning_rate": 9.620770128354726e-05,
154
+ "loss": 4.7012,
155
+ "step": 460
156
+ },
157
+ {
158
+ "epoch": 0.24,
159
+ "learning_rate": 9.604100683447241e-05,
160
+ "loss": 4.7355,
161
+ "step": 480
162
+ },
163
+ {
164
+ "epoch": 0.25,
165
+ "learning_rate": 9.587431238539756e-05,
166
+ "loss": 4.7008,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 0.26,
171
+ "learning_rate": 9.570761793632273e-05,
172
+ "loss": 4.6625,
173
+ "step": 520
174
+ },
175
+ {
176
+ "epoch": 0.27,
177
+ "learning_rate": 9.554092348724788e-05,
178
+ "loss": 4.6586,
179
+ "step": 540
180
+ },
181
+ {
182
+ "epoch": 0.28,
183
+ "learning_rate": 9.537422903817303e-05,
184
+ "loss": 4.6585,
185
+ "step": 560
186
+ },
187
+ {
188
+ "epoch": 0.29,
189
+ "learning_rate": 9.52075345890982e-05,
190
+ "loss": 4.6194,
191
+ "step": 580
192
+ },
193
+ {
194
+ "epoch": 0.3,
195
+ "learning_rate": 9.504084014002334e-05,
196
+ "loss": 4.6275,
197
+ "step": 600
198
+ },
199
+ {
200
+ "epoch": 0.31,
201
+ "learning_rate": 9.48741456909485e-05,
202
+ "loss": 4.6214,
203
+ "step": 620
204
+ },
205
+ {
206
+ "epoch": 0.32,
207
+ "learning_rate": 9.470745124187365e-05,
208
+ "loss": 4.6104,
209
+ "step": 640
210
+ },
211
+ {
212
+ "epoch": 0.33,
213
+ "learning_rate": 9.45407567927988e-05,
214
+ "loss": 4.5967,
215
+ "step": 660
216
+ },
217
+ {
218
+ "epoch": 0.34,
219
+ "learning_rate": 9.437406234372396e-05,
220
+ "loss": 4.568,
221
+ "step": 680
222
+ },
223
+ {
224
+ "epoch": 0.35,
225
+ "learning_rate": 9.420736789464911e-05,
226
+ "loss": 4.5526,
227
+ "step": 700
228
+ },
229
+ {
230
+ "epoch": 0.36,
231
+ "learning_rate": 9.404067344557426e-05,
232
+ "loss": 4.5289,
233
+ "step": 720
234
+ },
235
+ {
236
+ "epoch": 0.37,
237
+ "learning_rate": 9.387397899649943e-05,
238
+ "loss": 4.5548,
239
+ "step": 740
240
+ },
241
+ {
242
+ "epoch": 0.38,
243
+ "learning_rate": 9.370728454742458e-05,
244
+ "loss": 4.521,
245
+ "step": 760
246
+ },
247
+ {
248
+ "epoch": 0.39,
249
+ "learning_rate": 9.354059009834973e-05,
250
+ "loss": 4.5156,
251
+ "step": 780
252
+ },
253
+ {
254
+ "epoch": 0.4,
255
+ "learning_rate": 9.337389564927488e-05,
256
+ "loss": 4.5267,
257
+ "step": 800
258
+ },
259
+ {
260
+ "epoch": 0.4,
261
+ "eval_validation_accuracy": 0.6666666666666666,
262
+ "eval_validation_loss": 7.04296875,
263
+ "eval_validation_runtime": 332.7445,
264
+ "eval_validation_samples_per_second": 31.042,
265
+ "eval_validation_steps_per_second": 1.295,
266
+ "step": 800
267
+ },
268
+ {
269
+ "epoch": 0.41,
270
+ "learning_rate": 9.320720120020003e-05,
271
+ "loss": 4.479,
272
+ "step": 820
273
+ },
274
+ {
275
+ "epoch": 0.42,
276
+ "learning_rate": 9.304050675112518e-05,
277
+ "loss": 4.4737,
278
+ "step": 840
279
+ },
280
+ {
281
+ "epoch": 0.43,
282
+ "learning_rate": 9.287381230205035e-05,
283
+ "loss": 4.4788,
284
+ "step": 860
285
+ },
286
+ {
287
+ "epoch": 0.44,
288
+ "learning_rate": 9.27071178529755e-05,
289
+ "loss": 4.5016,
290
+ "step": 880
291
+ },
292
+ {
293
+ "epoch": 0.45,
294
+ "learning_rate": 9.254042340390065e-05,
295
+ "loss": 4.4504,
296
+ "step": 900
297
+ },
298
+ {
299
+ "epoch": 0.46,
300
+ "learning_rate": 9.237372895482581e-05,
301
+ "loss": 4.4458,
302
+ "step": 920
303
+ },
304
+ {
305
+ "epoch": 0.47,
306
+ "learning_rate": 9.220703450575097e-05,
307
+ "loss": 4.419,
308
+ "step": 940
309
+ },
310
+ {
311
+ "epoch": 0.48,
312
+ "learning_rate": 9.204034005667612e-05,
313
+ "loss": 4.4213,
314
+ "step": 960
315
+ },
316
+ {
317
+ "epoch": 0.49,
318
+ "learning_rate": 9.187364560760127e-05,
319
+ "loss": 4.4367,
320
+ "step": 980
321
+ },
322
+ {
323
+ "epoch": 0.5,
324
+ "learning_rate": 9.170695115852642e-05,
325
+ "loss": 4.4011,
326
+ "step": 1000
327
+ },
328
+ {
329
+ "epoch": 0.51,
330
+ "learning_rate": 9.155692615435907e-05,
331
+ "loss": 4.3795,
332
+ "step": 1020
333
+ },
334
+ {
335
+ "epoch": 0.52,
336
+ "learning_rate": 9.139023170528422e-05,
337
+ "loss": 4.4159,
338
+ "step": 1040
339
+ },
340
+ {
341
+ "epoch": 0.53,
342
+ "learning_rate": 9.122353725620937e-05,
343
+ "loss": 4.3601,
344
+ "step": 1060
345
+ },
346
+ {
347
+ "epoch": 0.54,
348
+ "learning_rate": 9.105684280713452e-05,
349
+ "loss": 4.368,
350
+ "step": 1080
351
+ },
352
+ {
353
+ "epoch": 0.55,
354
+ "learning_rate": 9.089014835805968e-05,
355
+ "loss": 4.3598,
356
+ "step": 1100
357
+ },
358
+ {
359
+ "epoch": 0.56,
360
+ "learning_rate": 9.072345390898484e-05,
361
+ "loss": 4.3406,
362
+ "step": 1120
363
+ },
364
+ {
365
+ "epoch": 0.57,
366
+ "learning_rate": 9.055675945990999e-05,
367
+ "loss": 4.3322,
368
+ "step": 1140
369
+ },
370
+ {
371
+ "epoch": 0.58,
372
+ "learning_rate": 9.039006501083514e-05,
373
+ "loss": 4.3484,
374
+ "step": 1160
375
+ },
376
+ {
377
+ "epoch": 0.59,
378
+ "learning_rate": 9.02233705617603e-05,
379
+ "loss": 4.3258,
380
+ "step": 1180
381
+ },
382
+ {
383
+ "epoch": 0.6,
384
+ "learning_rate": 9.005667611268546e-05,
385
+ "loss": 4.3159,
386
+ "step": 1200
387
+ },
388
+ {
389
+ "epoch": 0.6,
390
+ "eval_validation_accuracy": 0.6666666666666666,
391
+ "eval_validation_loss": 6.78125,
392
+ "eval_validation_runtime": 331.7984,
393
+ "eval_validation_samples_per_second": 31.13,
394
+ "eval_validation_steps_per_second": 1.299,
395
+ "step": 1200
396
+ },
397
+ {
398
+ "epoch": 0.61,
399
+ "learning_rate": 8.98899816636106e-05,
400
+ "loss": 4.3241,
401
+ "step": 1220
402
+ },
403
+ {
404
+ "epoch": 0.62,
405
+ "learning_rate": 8.972328721453576e-05,
406
+ "loss": 4.3158,
407
+ "step": 1240
408
+ },
409
+ {
410
+ "epoch": 0.63,
411
+ "learning_rate": 8.955659276546091e-05,
412
+ "loss": 4.2958,
413
+ "step": 1260
414
+ },
415
+ {
416
+ "epoch": 0.64,
417
+ "learning_rate": 8.938989831638606e-05,
418
+ "loss": 4.2935,
419
+ "step": 1280
420
+ },
421
+ {
422
+ "epoch": 0.65,
423
+ "learning_rate": 8.922320386731123e-05,
424
+ "loss": 4.3085,
425
+ "step": 1300
426
+ },
427
+ {
428
+ "epoch": 0.66,
429
+ "learning_rate": 8.905650941823638e-05,
430
+ "loss": 4.2939,
431
+ "step": 1320
432
+ },
433
+ {
434
+ "epoch": 0.67,
435
+ "learning_rate": 8.888981496916153e-05,
436
+ "loss": 4.3005,
437
+ "step": 1340
438
+ },
439
+ {
440
+ "epoch": 0.68,
441
+ "learning_rate": 8.872312052008669e-05,
442
+ "loss": 4.2749,
443
+ "step": 1360
444
+ },
445
+ {
446
+ "epoch": 0.69,
447
+ "learning_rate": 8.855642607101184e-05,
448
+ "loss": 4.2629,
449
+ "step": 1380
450
+ },
451
+ {
452
+ "epoch": 0.7,
453
+ "learning_rate": 8.8389731621937e-05,
454
+ "loss": 4.2208,
455
+ "step": 1400
456
+ },
457
+ {
458
+ "epoch": 0.71,
459
+ "learning_rate": 8.822303717286215e-05,
460
+ "loss": 4.2333,
461
+ "step": 1420
462
+ },
463
+ {
464
+ "epoch": 0.72,
465
+ "learning_rate": 8.80563427237873e-05,
466
+ "loss": 4.2281,
467
+ "step": 1440
468
+ },
469
+ {
470
+ "epoch": 0.73,
471
+ "learning_rate": 8.788964827471246e-05,
472
+ "loss": 4.2151,
473
+ "step": 1460
474
+ },
475
+ {
476
+ "epoch": 0.74,
477
+ "learning_rate": 8.772295382563761e-05,
478
+ "loss": 4.2278,
479
+ "step": 1480
480
+ },
481
+ {
482
+ "epoch": 0.75,
483
+ "learning_rate": 8.755625937656276e-05,
484
+ "loss": 4.1744,
485
+ "step": 1500
486
+ },
487
+ {
488
+ "epoch": 0.76,
489
+ "learning_rate": 8.738956492748793e-05,
490
+ "loss": 4.2359,
491
+ "step": 1520
492
+ },
493
+ {
494
+ "epoch": 0.77,
495
+ "learning_rate": 8.722287047841308e-05,
496
+ "loss": 4.1841,
497
+ "step": 1540
498
+ },
499
+ {
500
+ "epoch": 0.78,
501
+ "learning_rate": 8.705617602933823e-05,
502
+ "loss": 4.1512,
503
+ "step": 1560
504
+ },
505
+ {
506
+ "epoch": 0.79,
507
+ "learning_rate": 8.688948158026338e-05,
508
+ "loss": 4.1918,
509
+ "step": 1580
510
+ },
511
+ {
512
+ "epoch": 0.8,
513
+ "learning_rate": 8.672278713118853e-05,
514
+ "loss": 4.1662,
515
+ "step": 1600
516
+ },
517
+ {
518
+ "epoch": 0.8,
519
+ "eval_validation_accuracy": 1.0,
520
+ "eval_validation_loss": 6.546875,
521
+ "eval_validation_runtime": 332.9747,
522
+ "eval_validation_samples_per_second": 31.02,
523
+ "eval_validation_steps_per_second": 1.294,
524
+ "step": 1600
525
+ },
526
+ {
527
+ "epoch": 0.81,
528
+ "learning_rate": 8.655609268211368e-05,
529
+ "loss": 4.1917,
530
+ "step": 1620
531
+ },
532
+ {
533
+ "epoch": 0.82,
534
+ "learning_rate": 8.638939823303885e-05,
535
+ "loss": 4.1628,
536
+ "step": 1640
537
+ },
538
+ {
539
+ "epoch": 0.83,
540
+ "learning_rate": 8.6222703783964e-05,
541
+ "loss": 4.1672,
542
+ "step": 1660
543
+ },
544
+ {
545
+ "epoch": 0.84,
546
+ "learning_rate": 8.605600933488915e-05,
547
+ "loss": 4.1224,
548
+ "step": 1680
549
+ },
550
+ {
551
+ "epoch": 0.85,
552
+ "learning_rate": 8.588931488581431e-05,
553
+ "loss": 4.1282,
554
+ "step": 1700
555
+ },
556
+ {
557
+ "epoch": 0.86,
558
+ "learning_rate": 8.572262043673946e-05,
559
+ "loss": 4.15,
560
+ "step": 1720
561
+ },
562
+ {
563
+ "epoch": 0.87,
564
+ "learning_rate": 8.555592598766461e-05,
565
+ "loss": 4.1399,
566
+ "step": 1740
567
+ },
568
+ {
569
+ "epoch": 0.88,
570
+ "learning_rate": 8.538923153858977e-05,
571
+ "loss": 4.1007,
572
+ "step": 1760
573
+ },
574
+ {
575
+ "epoch": 0.89,
576
+ "learning_rate": 8.522253708951492e-05,
577
+ "loss": 4.0977,
578
+ "step": 1780
579
+ },
580
+ {
581
+ "epoch": 0.9,
582
+ "learning_rate": 8.505584264044008e-05,
583
+ "loss": 4.0753,
584
+ "step": 1800
585
+ },
586
+ {
587
+ "epoch": 0.91,
588
+ "learning_rate": 8.488914819136523e-05,
589
+ "loss": 4.0965,
590
+ "step": 1820
591
+ },
592
+ {
593
+ "epoch": 0.92,
594
+ "learning_rate": 8.472245374229038e-05,
595
+ "loss": 4.0711,
596
+ "step": 1840
597
+ },
598
+ {
599
+ "epoch": 0.93,
600
+ "learning_rate": 8.455575929321555e-05,
601
+ "loss": 4.0451,
602
+ "step": 1860
603
+ },
604
+ {
605
+ "epoch": 0.94,
606
+ "learning_rate": 8.43890648441407e-05,
607
+ "loss": 4.0614,
608
+ "step": 1880
609
+ },
610
+ {
611
+ "epoch": 0.95,
612
+ "learning_rate": 8.422237039506585e-05,
613
+ "loss": 4.0577,
614
+ "step": 1900
615
+ },
616
+ {
617
+ "epoch": 0.96,
618
+ "learning_rate": 8.4055675945991e-05,
619
+ "loss": 4.0269,
620
+ "step": 1920
621
+ },
622
+ {
623
+ "epoch": 0.97,
624
+ "learning_rate": 8.388898149691615e-05,
625
+ "loss": 4.0438,
626
+ "step": 1940
627
+ },
628
+ {
629
+ "epoch": 0.98,
630
+ "learning_rate": 8.37222870478413e-05,
631
+ "loss": 4.0236,
632
+ "step": 1960
633
+ },
634
+ {
635
+ "epoch": 0.99,
636
+ "learning_rate": 8.355559259876647e-05,
637
+ "loss": 4.0584,
638
+ "step": 1980
639
+ },
640
+ {
641
+ "epoch": 1.0,
642
+ "learning_rate": 8.338889814969162e-05,
643
+ "loss": 4.0047,
644
+ "step": 2000
645
+ },
646
+ {
647
+ "epoch": 1.0,
648
+ "eval_validation_accuracy": 1.0,
649
+ "eval_validation_loss": 6.3046875,
650
+ "eval_validation_runtime": 330.2,
651
+ "eval_validation_samples_per_second": 31.281,
652
+ "eval_validation_steps_per_second": 1.305,
653
+ "step": 2000
654
+ },
655
+ {
656
+ "epoch": 1.01,
657
+ "learning_rate": 8.323887314552426e-05,
658
+ "loss": 4.269,
659
+ "step": 2020
660
+ },
661
+ {
662
+ "epoch": 1.02,
663
+ "learning_rate": 8.307217869644941e-05,
664
+ "loss": 4.0004,
665
+ "step": 2040
666
+ },
667
+ {
668
+ "epoch": 1.03,
669
+ "learning_rate": 8.290548424737456e-05,
670
+ "loss": 4.0091,
671
+ "step": 2060
672
+ },
673
+ {
674
+ "epoch": 1.04,
675
+ "learning_rate": 8.273878979829972e-05,
676
+ "loss": 3.9829,
677
+ "step": 2080
678
+ },
679
+ {
680
+ "epoch": 1.05,
681
+ "learning_rate": 8.257209534922488e-05,
682
+ "loss": 3.9605,
683
+ "step": 2100
684
+ },
685
+ {
686
+ "epoch": 1.06,
687
+ "learning_rate": 8.240540090015003e-05,
688
+ "loss": 3.9552,
689
+ "step": 2120
690
+ },
691
+ {
692
+ "epoch": 1.07,
693
+ "learning_rate": 8.223870645107519e-05,
694
+ "loss": 3.9393,
695
+ "step": 2140
696
+ },
697
+ {
698
+ "epoch": 1.08,
699
+ "learning_rate": 8.207201200200034e-05,
700
+ "loss": 3.9179,
701
+ "step": 2160
702
+ },
703
+ {
704
+ "epoch": 1.09,
705
+ "learning_rate": 8.190531755292549e-05,
706
+ "loss": 3.9201,
707
+ "step": 2180
708
+ },
709
+ {
710
+ "epoch": 1.1,
711
+ "learning_rate": 8.173862310385064e-05,
712
+ "loss": 3.9166,
713
+ "step": 2200
714
+ },
715
+ {
716
+ "epoch": 1.11,
717
+ "learning_rate": 8.15719286547758e-05,
718
+ "loss": 3.9005,
719
+ "step": 2220
720
+ },
721
+ {
722
+ "epoch": 1.12,
723
+ "learning_rate": 8.140523420570096e-05,
724
+ "loss": 3.9457,
725
+ "step": 2240
726
+ },
727
+ {
728
+ "epoch": 1.13,
729
+ "learning_rate": 8.123853975662611e-05,
730
+ "loss": 3.8752,
731
+ "step": 2260
732
+ },
733
+ {
734
+ "epoch": 1.14,
735
+ "learning_rate": 8.107184530755126e-05,
736
+ "loss": 3.8767,
737
+ "step": 2280
738
+ },
739
+ {
740
+ "epoch": 1.15,
741
+ "learning_rate": 8.090515085847643e-05,
742
+ "loss": 3.8952,
743
+ "step": 2300
744
+ },
745
+ {
746
+ "epoch": 1.16,
747
+ "learning_rate": 8.073845640940158e-05,
748
+ "loss": 3.9119,
749
+ "step": 2320
750
+ },
751
+ {
752
+ "epoch": 1.17,
753
+ "learning_rate": 8.057176196032673e-05,
754
+ "loss": 3.8805,
755
+ "step": 2340
756
+ },
757
+ {
758
+ "epoch": 1.18,
759
+ "learning_rate": 8.040506751125188e-05,
760
+ "loss": 3.8966,
761
+ "step": 2360
762
+ },
763
+ {
764
+ "epoch": 1.19,
765
+ "learning_rate": 8.023837306217703e-05,
766
+ "loss": 3.8415,
767
+ "step": 2380
768
+ },
769
+ {
770
+ "epoch": 1.2,
771
+ "learning_rate": 8.007167861310218e-05,
772
+ "loss": 3.8951,
773
+ "step": 2400
774
+ },
775
+ {
776
+ "epoch": 1.2,
777
+ "eval_validation_accuracy": 1.0,
778
+ "eval_validation_loss": 6.078125,
779
+ "eval_validation_runtime": 332.7355,
780
+ "eval_validation_samples_per_second": 31.043,
781
+ "eval_validation_steps_per_second": 1.295,
782
+ "step": 2400
783
+ },
784
+ {
785
+ "epoch": 1.21,
786
+ "learning_rate": 7.990498416402735e-05,
787
+ "loss": 3.8927,
788
+ "step": 2420
789
+ },
790
+ {
791
+ "epoch": 1.22,
792
+ "learning_rate": 7.97382897149525e-05,
793
+ "loss": 3.8585,
794
+ "step": 2440
795
+ },
796
+ {
797
+ "epoch": 1.23,
798
+ "learning_rate": 7.957159526587765e-05,
799
+ "loss": 3.8454,
800
+ "step": 2460
801
+ },
802
+ {
803
+ "epoch": 1.24,
804
+ "learning_rate": 7.940490081680281e-05,
805
+ "loss": 3.8487,
806
+ "step": 2480
807
+ },
808
+ {
809
+ "epoch": 1.25,
810
+ "learning_rate": 7.923820636772796e-05,
811
+ "loss": 3.864,
812
+ "step": 2500
813
+ },
814
+ {
815
+ "epoch": 1.26,
816
+ "learning_rate": 7.907151191865311e-05,
817
+ "loss": 5.1011,
818
+ "step": 2520
819
+ },
820
+ {
821
+ "epoch": 1.27,
822
+ "learning_rate": 7.890481746957826e-05,
823
+ "loss": 5.0875,
824
+ "step": 2540
825
+ },
826
+ {
827
+ "epoch": 1.28,
828
+ "learning_rate": 7.873812302050342e-05,
829
+ "loss": 5.0821,
830
+ "step": 2560
831
+ },
832
+ {
833
+ "epoch": 1.29,
834
+ "learning_rate": 7.857142857142858e-05,
835
+ "loss": 5.1312,
836
+ "step": 2580
837
+ },
838
+ {
839
+ "epoch": 1.3,
840
+ "learning_rate": 7.840473412235373e-05,
841
+ "loss": 5.0985,
842
+ "step": 2600
843
+ },
844
+ {
845
+ "epoch": 1.31,
846
+ "learning_rate": 7.823803967327888e-05,
847
+ "loss": 5.0481,
848
+ "step": 2620
849
+ },
850
+ {
851
+ "epoch": 1.32,
852
+ "learning_rate": 7.807134522420405e-05,
853
+ "loss": 5.1117,
854
+ "step": 2640
855
+ },
856
+ {
857
+ "epoch": 1.33,
858
+ "learning_rate": 7.79046507751292e-05,
859
+ "loss": 5.0702,
860
+ "step": 2660
861
+ },
862
+ {
863
+ "epoch": 1.34,
864
+ "learning_rate": 7.773795632605435e-05,
865
+ "loss": 4.9993,
866
+ "step": 2680
867
+ },
868
+ {
869
+ "epoch": 1.35,
870
+ "learning_rate": 7.75712618769795e-05,
871
+ "loss": 5.0139,
872
+ "step": 2700
873
+ },
874
+ {
875
+ "epoch": 1.36,
876
+ "learning_rate": 7.740456742790465e-05,
877
+ "loss": 5.0058,
878
+ "step": 2720
879
+ },
880
+ {
881
+ "epoch": 1.37,
882
+ "learning_rate": 7.72378729788298e-05,
883
+ "loss": 5.03,
884
+ "step": 2740
885
+ },
886
+ {
887
+ "epoch": 1.38,
888
+ "learning_rate": 7.707117852975497e-05,
889
+ "loss": 5.0385,
890
+ "step": 2760
891
+ },
892
+ {
893
+ "epoch": 1.39,
894
+ "learning_rate": 7.690448408068012e-05,
895
+ "loss": 5.0209,
896
+ "step": 2780
897
+ },
898
+ {
899
+ "epoch": 1.4,
900
+ "learning_rate": 7.673778963160527e-05,
901
+ "loss": 5.0055,
902
+ "step": 2800
903
+ },
904
+ {
905
+ "epoch": 1.4,
906
+ "eval_validation_accuracy": 1.0,
907
+ "eval_validation_loss": 5.8984375,
908
+ "eval_validation_runtime": 337.5487,
909
+ "eval_validation_samples_per_second": 30.6,
910
+ "eval_validation_steps_per_second": 1.277,
911
+ "step": 2800
912
+ },
913
+ {
914
+ "epoch": 1.41,
915
+ "learning_rate": 7.657109518253043e-05,
916
+ "loss": 5.0003,
917
+ "step": 2820
918
+ },
919
+ {
920
+ "epoch": 1.42,
921
+ "learning_rate": 7.640440073345558e-05,
922
+ "loss": 4.9803,
923
+ "step": 2840
924
+ },
925
+ {
926
+ "epoch": 1.43,
927
+ "learning_rate": 7.623770628438073e-05,
928
+ "loss": 4.9668,
929
+ "step": 2860
930
+ },
931
+ {
932
+ "epoch": 1.44,
933
+ "learning_rate": 7.607101183530589e-05,
934
+ "loss": 4.9738,
935
+ "step": 2880
936
+ },
937
+ {
938
+ "epoch": 1.45,
939
+ "learning_rate": 7.590431738623104e-05,
940
+ "loss": 4.949,
941
+ "step": 2900
942
+ },
943
+ {
944
+ "epoch": 1.46,
945
+ "learning_rate": 7.57376229371562e-05,
946
+ "loss": 4.9473,
947
+ "step": 2920
948
+ },
949
+ {
950
+ "epoch": 1.47,
951
+ "learning_rate": 7.557092848808135e-05,
952
+ "loss": 4.9476,
953
+ "step": 2940
954
+ },
955
+ {
956
+ "epoch": 1.48,
957
+ "learning_rate": 7.54042340390065e-05,
958
+ "loss": 4.9322,
959
+ "step": 2960
960
+ },
961
+ {
962
+ "epoch": 1.49,
963
+ "learning_rate": 7.523753958993165e-05,
964
+ "loss": 4.9849,
965
+ "step": 2980
966
+ },
967
+ {
968
+ "epoch": 1.5,
969
+ "learning_rate": 7.507084514085682e-05,
970
+ "loss": 4.8861,
971
+ "step": 3000
972
+ },
973
+ {
974
+ "epoch": 1.51,
975
+ "learning_rate": 7.492082013668946e-05,
976
+ "loss": 4.9622,
977
+ "step": 3020
978
+ },
979
+ {
980
+ "epoch": 1.52,
981
+ "learning_rate": 7.475412568761461e-05,
982
+ "loss": 4.9207,
983
+ "step": 3040
984
+ },
985
+ {
986
+ "epoch": 1.53,
987
+ "learning_rate": 7.458743123853976e-05,
988
+ "loss": 4.9588,
989
+ "step": 3060
990
+ },
991
+ {
992
+ "epoch": 1.54,
993
+ "learning_rate": 7.442073678946491e-05,
994
+ "loss": 4.9494,
995
+ "step": 3080
996
+ },
997
+ {
998
+ "epoch": 1.55,
999
+ "learning_rate": 7.425404234039008e-05,
1000
+ "loss": 4.9338,
1001
+ "step": 3100
1002
+ },
1003
+ {
1004
+ "epoch": 1.56,
1005
+ "learning_rate": 7.408734789131521e-05,
1006
+ "loss": 4.9022,
1007
+ "step": 3120
1008
+ },
1009
+ {
1010
+ "epoch": 1.57,
1011
+ "learning_rate": 7.392065344224038e-05,
1012
+ "loss": 4.9907,
1013
+ "step": 3140
1014
+ },
1015
+ {
1016
+ "epoch": 1.58,
1017
+ "learning_rate": 7.375395899316553e-05,
1018
+ "loss": 4.9025,
1019
+ "step": 3160
1020
+ },
1021
+ {
1022
+ "epoch": 1.59,
1023
+ "learning_rate": 7.358726454409068e-05,
1024
+ "loss": 4.8806,
1025
+ "step": 3180
1026
+ },
1027
+ {
1028
+ "epoch": 1.6,
1029
+ "learning_rate": 7.342057009501584e-05,
1030
+ "loss": 4.9251,
1031
+ "step": 3200
1032
+ },
1033
+ {
1034
+ "epoch": 1.6,
1035
+ "eval_validation_accuracy": 1.0,
1036
+ "eval_validation_loss": 5.76171875,
1037
+ "eval_validation_runtime": 336.3444,
1038
+ "eval_validation_samples_per_second": 30.71,
1039
+ "eval_validation_steps_per_second": 1.281,
1040
+ "step": 3200
1041
+ },
1042
+ {
1043
+ "epoch": 1.61,
1044
+ "learning_rate": 7.3253875645941e-05,
1045
+ "loss": 4.8988,
1046
+ "step": 3220
1047
+ },
1048
+ {
1049
+ "epoch": 1.62,
1050
+ "learning_rate": 7.308718119686615e-05,
1051
+ "loss": 4.9339,
1052
+ "step": 3240
1053
+ },
1054
+ {
1055
+ "epoch": 1.63,
1056
+ "learning_rate": 7.292048674779131e-05,
1057
+ "loss": 4.8793,
1058
+ "step": 3260
1059
+ },
1060
+ {
1061
+ "epoch": 1.64,
1062
+ "learning_rate": 7.275379229871646e-05,
1063
+ "loss": 4.8935,
1064
+ "step": 3280
1065
+ },
1066
+ {
1067
+ "epoch": 1.65,
1068
+ "learning_rate": 7.25870978496416e-05,
1069
+ "loss": 4.8633,
1070
+ "step": 3300
1071
+ },
1072
+ {
1073
+ "epoch": 1.66,
1074
+ "learning_rate": 7.242040340056676e-05,
1075
+ "loss": 4.8315,
1076
+ "step": 3320
1077
+ },
1078
+ {
1079
+ "epoch": 1.67,
1080
+ "learning_rate": 7.225370895149191e-05,
1081
+ "loss": 4.8529,
1082
+ "step": 3340
1083
+ },
1084
+ {
1085
+ "epoch": 1.67,
1086
+ "learning_rate": 7.208701450241707e-05,
1087
+ "loss": 4.8308,
1088
+ "step": 3360
1089
+ },
1090
+ {
1091
+ "epoch": 1.68,
1092
+ "learning_rate": 7.192032005334223e-05,
1093
+ "loss": 4.8546,
1094
+ "step": 3380
1095
+ },
1096
+ {
1097
+ "epoch": 1.69,
1098
+ "learning_rate": 7.175362560426738e-05,
1099
+ "loss": 4.8793,
1100
+ "step": 3400
1101
+ },
1102
+ {
1103
+ "epoch": 1.7,
1104
+ "learning_rate": 7.158693115519253e-05,
1105
+ "loss": 4.85,
1106
+ "step": 3420
1107
+ },
1108
+ {
1109
+ "epoch": 1.71,
1110
+ "learning_rate": 7.14202367061177e-05,
1111
+ "loss": 4.8368,
1112
+ "step": 3440
1113
+ },
1114
+ {
1115
+ "epoch": 1.72,
1116
+ "learning_rate": 7.125354225704285e-05,
1117
+ "loss": 4.8551,
1118
+ "step": 3460
1119
+ },
1120
+ {
1121
+ "epoch": 1.73,
1122
+ "learning_rate": 7.1086847807968e-05,
1123
+ "loss": 4.8289,
1124
+ "step": 3480
1125
+ },
1126
+ {
1127
+ "epoch": 1.74,
1128
+ "learning_rate": 7.092015335889315e-05,
1129
+ "loss": 4.86,
1130
+ "step": 3500
1131
+ },
1132
+ {
1133
+ "epoch": 1.75,
1134
+ "learning_rate": 7.07534589098183e-05,
1135
+ "loss": 4.8223,
1136
+ "step": 3520
1137
+ },
1138
+ {
1139
+ "epoch": 1.76,
1140
+ "learning_rate": 7.058676446074346e-05,
1141
+ "loss": 4.84,
1142
+ "step": 3540
1143
+ },
1144
+ {
1145
+ "epoch": 1.77,
1146
+ "learning_rate": 7.042007001166862e-05,
1147
+ "loss": 4.7863,
1148
+ "step": 3560
1149
+ },
1150
+ {
1151
+ "epoch": 1.78,
1152
+ "learning_rate": 7.025337556259377e-05,
1153
+ "loss": 4.8636,
1154
+ "step": 3580
1155
+ },
1156
+ {
1157
+ "epoch": 1.79,
1158
+ "learning_rate": 7.008668111351893e-05,
1159
+ "loss": 4.8217,
1160
+ "step": 3600
1161
+ },
1162
+ {
1163
+ "epoch": 1.79,
1164
+ "eval_validation_accuracy": 1.0,
1165
+ "eval_validation_loss": 5.65625,
1166
+ "eval_validation_runtime": 340.8771,
1167
+ "eval_validation_samples_per_second": 30.301,
1168
+ "eval_validation_steps_per_second": 1.264,
1169
+ "step": 3600
1170
+ },
1171
+ {
1172
+ "epoch": 1.8,
1173
+ "learning_rate": 6.991998666444408e-05,
1174
+ "loss": 4.7671,
1175
+ "step": 3620
1176
+ },
1177
+ {
1178
+ "epoch": 1.81,
1179
+ "learning_rate": 6.975329221536923e-05,
1180
+ "loss": 4.8273,
1181
+ "step": 3640
1182
+ },
1183
+ {
1184
+ "epoch": 1.82,
1185
+ "learning_rate": 6.958659776629438e-05,
1186
+ "loss": 4.845,
1187
+ "step": 3660
1188
+ },
1189
+ {
1190
+ "epoch": 1.83,
1191
+ "learning_rate": 6.941990331721953e-05,
1192
+ "loss": 4.7974,
1193
+ "step": 3680
1194
+ },
1195
+ {
1196
+ "epoch": 1.84,
1197
+ "learning_rate": 6.925320886814469e-05,
1198
+ "loss": 4.7464,
1199
+ "step": 3700
1200
+ },
1201
+ {
1202
+ "epoch": 1.85,
1203
+ "learning_rate": 6.908651441906985e-05,
1204
+ "loss": 4.8007,
1205
+ "step": 3720
1206
+ },
1207
+ {
1208
+ "epoch": 1.86,
1209
+ "learning_rate": 6.8919819969995e-05,
1210
+ "loss": 4.72,
1211
+ "step": 3740
1212
+ },
1213
+ {
1214
+ "epoch": 1.87,
1215
+ "learning_rate": 6.875312552092015e-05,
1216
+ "loss": 4.7967,
1217
+ "step": 3760
1218
+ },
1219
+ {
1220
+ "epoch": 1.88,
1221
+ "learning_rate": 6.858643107184532e-05,
1222
+ "loss": 4.807,
1223
+ "step": 3780
1224
+ },
1225
+ {
1226
+ "epoch": 1.89,
1227
+ "learning_rate": 6.841973662277047e-05,
1228
+ "loss": 4.7475,
1229
+ "step": 3800
1230
+ },
1231
+ {
1232
+ "epoch": 1.9,
1233
+ "learning_rate": 6.825304217369562e-05,
1234
+ "loss": 4.7911,
1235
+ "step": 3820
1236
+ },
1237
+ {
1238
+ "epoch": 1.91,
1239
+ "learning_rate": 6.808634772462077e-05,
1240
+ "loss": 4.7981,
1241
+ "step": 3840
1242
+ },
1243
+ {
1244
+ "epoch": 1.92,
1245
+ "learning_rate": 6.791965327554592e-05,
1246
+ "loss": 4.7561,
1247
+ "step": 3860
1248
+ },
1249
+ {
1250
+ "epoch": 1.93,
1251
+ "learning_rate": 6.775295882647109e-05,
1252
+ "loss": 4.7958,
1253
+ "step": 3880
1254
+ },
1255
+ {
1256
+ "epoch": 1.94,
1257
+ "learning_rate": 6.758626437739624e-05,
1258
+ "loss": 4.7506,
1259
+ "step": 3900
1260
+ },
1261
+ {
1262
+ "epoch": 1.95,
1263
+ "learning_rate": 6.741956992832139e-05,
1264
+ "loss": 4.7785,
1265
+ "step": 3920
1266
+ },
1267
+ {
1268
+ "epoch": 1.96,
1269
+ "learning_rate": 6.725287547924655e-05,
1270
+ "loss": 4.7323,
1271
+ "step": 3940
1272
+ },
1273
+ {
1274
+ "epoch": 1.97,
1275
+ "learning_rate": 6.70861810301717e-05,
1276
+ "loss": 4.7261,
1277
+ "step": 3960
1278
+ },
1279
+ {
1280
+ "epoch": 1.98,
1281
+ "learning_rate": 6.691948658109685e-05,
1282
+ "loss": 4.7176,
1283
+ "step": 3980
1284
+ },
1285
+ {
1286
+ "epoch": 1.99,
1287
+ "learning_rate": 6.6752792132022e-05,
1288
+ "loss": 4.7685,
1289
+ "step": 4000
1290
+ },
1291
+ {
1292
+ "epoch": 1.99,
1293
+ "eval_validation_accuracy": 1.0,
1294
+ "eval_validation_loss": 5.56640625,
1295
+ "eval_validation_runtime": 337.8469,
1296
+ "eval_validation_samples_per_second": 30.573,
1297
+ "eval_validation_steps_per_second": 1.276,
1298
+ "step": 4000
1299
+ },
1300
+ {
1301
+ "epoch": 2.0,
1302
+ "learning_rate": 6.660276712785464e-05,
1303
+ "loss": 4.7444,
1304
+ "step": 4020
1305
+ },
1306
+ {
1307
+ "epoch": 2.01,
1308
+ "learning_rate": 6.643607267877981e-05,
1309
+ "loss": 4.6876,
1310
+ "step": 4040
1311
+ },
1312
+ {
1313
+ "epoch": 2.02,
1314
+ "learning_rate": 6.626937822970496e-05,
1315
+ "loss": 4.7004,
1316
+ "step": 4060
1317
+ },
1318
+ {
1319
+ "epoch": 2.03,
1320
+ "learning_rate": 6.61026837806301e-05,
1321
+ "loss": 4.6872,
1322
+ "step": 4080
1323
+ },
1324
+ {
1325
+ "epoch": 2.04,
1326
+ "learning_rate": 6.593598933155526e-05,
1327
+ "loss": 4.6648,
1328
+ "step": 4100
1329
+ },
1330
+ {
1331
+ "epoch": 2.05,
1332
+ "learning_rate": 6.576929488248041e-05,
1333
+ "loss": 4.6547,
1334
+ "step": 4120
1335
+ },
1336
+ {
1337
+ "epoch": 2.06,
1338
+ "learning_rate": 6.560260043340556e-05,
1339
+ "loss": 4.6732,
1340
+ "step": 4140
1341
+ },
1342
+ {
1343
+ "epoch": 2.07,
1344
+ "learning_rate": 6.543590598433073e-05,
1345
+ "loss": 4.6572,
1346
+ "step": 4160
1347
+ },
1348
+ {
1349
+ "epoch": 2.08,
1350
+ "learning_rate": 6.526921153525588e-05,
1351
+ "loss": 4.6615,
1352
+ "step": 4180
1353
+ },
1354
+ {
1355
+ "epoch": 2.09,
1356
+ "learning_rate": 6.510251708618103e-05,
1357
+ "loss": 4.6894,
1358
+ "step": 4200
1359
+ },
1360
+ {
1361
+ "epoch": 2.1,
1362
+ "learning_rate": 6.49358226371062e-05,
1363
+ "loss": 4.6807,
1364
+ "step": 4220
1365
+ },
1366
+ {
1367
+ "epoch": 2.11,
1368
+ "learning_rate": 6.476912818803135e-05,
1369
+ "loss": 4.6923,
1370
+ "step": 4240
1371
+ },
1372
+ {
1373
+ "epoch": 2.12,
1374
+ "learning_rate": 6.46024337389565e-05,
1375
+ "loss": 4.6416,
1376
+ "step": 4260
1377
+ },
1378
+ {
1379
+ "epoch": 2.13,
1380
+ "learning_rate": 6.443573928988165e-05,
1381
+ "loss": 4.6612,
1382
+ "step": 4280
1383
+ },
1384
+ {
1385
+ "epoch": 2.14,
1386
+ "learning_rate": 6.42690448408068e-05,
1387
+ "loss": 4.7298,
1388
+ "step": 4300
1389
+ },
1390
+ {
1391
+ "epoch": 2.15,
1392
+ "learning_rate": 6.410235039173196e-05,
1393
+ "loss": 4.6681,
1394
+ "step": 4320
1395
+ },
1396
+ {
1397
+ "epoch": 2.16,
1398
+ "learning_rate": 6.393565594265711e-05,
1399
+ "loss": 4.6858,
1400
+ "step": 4340
1401
+ },
1402
+ {
1403
+ "epoch": 2.17,
1404
+ "learning_rate": 6.376896149358227e-05,
1405
+ "loss": 4.6812,
1406
+ "step": 4360
1407
+ },
1408
+ {
1409
+ "epoch": 2.18,
1410
+ "learning_rate": 6.360226704450743e-05,
1411
+ "loss": 4.6864,
1412
+ "step": 4380
1413
+ },
1414
+ {
1415
+ "epoch": 2.19,
1416
+ "learning_rate": 6.343557259543258e-05,
1417
+ "loss": 4.6598,
1418
+ "step": 4400
1419
+ },
1420
+ {
1421
+ "epoch": 2.19,
1422
+ "eval_validation_accuracy": 1.0,
1423
+ "eval_validation_loss": 5.49609375,
1424
+ "eval_validation_runtime": 339.8448,
1425
+ "eval_validation_samples_per_second": 30.393,
1426
+ "eval_validation_steps_per_second": 1.268,
1427
+ "step": 4400
1428
+ },
1429
+ {
1430
+ "epoch": 2.2,
1431
+ "learning_rate": 6.326887814635772e-05,
1432
+ "loss": 4.6622,
1433
+ "step": 4420
1434
+ },
1435
+ {
1436
+ "epoch": 2.21,
1437
+ "learning_rate": 6.310218369728288e-05,
1438
+ "loss": 4.6558,
1439
+ "step": 4440
1440
+ },
1441
+ {
1442
+ "epoch": 2.22,
1443
+ "learning_rate": 6.293548924820803e-05,
1444
+ "loss": 4.6826,
1445
+ "step": 4460
1446
+ },
1447
+ {
1448
+ "epoch": 2.23,
1449
+ "learning_rate": 6.276879479913318e-05,
1450
+ "loss": 4.6494,
1451
+ "step": 4480
1452
+ },
1453
+ {
1454
+ "epoch": 2.24,
1455
+ "learning_rate": 6.260210035005835e-05,
1456
+ "loss": 4.6403,
1457
+ "step": 4500
1458
+ },
1459
+ {
1460
+ "epoch": 2.25,
1461
+ "learning_rate": 6.24354059009835e-05,
1462
+ "loss": 4.6279,
1463
+ "step": 4520
1464
+ },
1465
+ {
1466
+ "epoch": 2.26,
1467
+ "learning_rate": 6.226871145190865e-05,
1468
+ "loss": 4.634,
1469
+ "step": 4540
1470
+ },
1471
+ {
1472
+ "epoch": 2.27,
1473
+ "learning_rate": 6.210201700283382e-05,
1474
+ "loss": 4.6027,
1475
+ "step": 4560
1476
+ },
1477
+ {
1478
+ "epoch": 2.28,
1479
+ "learning_rate": 6.193532255375897e-05,
1480
+ "loss": 4.6261,
1481
+ "step": 4580
1482
+ },
1483
+ {
1484
+ "epoch": 2.29,
1485
+ "learning_rate": 6.176862810468412e-05,
1486
+ "loss": 4.6151,
1487
+ "step": 4600
1488
+ },
1489
+ {
1490
+ "epoch": 2.3,
1491
+ "learning_rate": 6.160193365560927e-05,
1492
+ "loss": 4.637,
1493
+ "step": 4620
1494
+ },
1495
+ {
1496
+ "epoch": 2.31,
1497
+ "learning_rate": 6.143523920653442e-05,
1498
+ "loss": 4.6628,
1499
+ "step": 4640
1500
+ },
1501
+ {
1502
+ "epoch": 2.32,
1503
+ "learning_rate": 6.126854475745958e-05,
1504
+ "loss": 4.6502,
1505
+ "step": 4660
1506
+ },
1507
+ {
1508
+ "epoch": 2.33,
1509
+ "learning_rate": 6.110185030838474e-05,
1510
+ "loss": 4.6172,
1511
+ "step": 4680
1512
+ },
1513
+ {
1514
+ "epoch": 2.34,
1515
+ "learning_rate": 6.0935155859309886e-05,
1516
+ "loss": 4.5895,
1517
+ "step": 4700
1518
+ },
1519
+ {
1520
+ "epoch": 2.35,
1521
+ "learning_rate": 6.0768461410235044e-05,
1522
+ "loss": 4.5621,
1523
+ "step": 4720
1524
+ },
1525
+ {
1526
+ "epoch": 2.36,
1527
+ "learning_rate": 6.06017669611602e-05,
1528
+ "loss": 4.6395,
1529
+ "step": 4740
1530
+ },
1531
+ {
1532
+ "epoch": 2.37,
1533
+ "learning_rate": 6.043507251208535e-05,
1534
+ "loss": 4.6082,
1535
+ "step": 4760
1536
+ },
1537
+ {
1538
+ "epoch": 2.38,
1539
+ "learning_rate": 6.02683780630105e-05,
1540
+ "loss": 4.6393,
1541
+ "step": 4780
1542
+ },
1543
+ {
1544
+ "epoch": 2.39,
1545
+ "learning_rate": 6.0101683613935654e-05,
1546
+ "loss": 4.6418,
1547
+ "step": 4800
1548
+ },
1549
+ {
1550
+ "epoch": 2.39,
1551
+ "eval_validation_accuracy": 1.0,
1552
+ "eval_validation_loss": 5.43359375,
1553
+ "eval_validation_runtime": 339.9919,
1554
+ "eval_validation_samples_per_second": 30.38,
1555
+ "eval_validation_steps_per_second": 1.268,
1556
+ "step": 4800
1557
+ },
1558
+ {
1559
+ "epoch": 2.4,
1560
+ "learning_rate": 5.993498916486081e-05,
1561
+ "loss": 4.6027,
1562
+ "step": 4820
1563
+ },
1564
+ {
1565
+ "epoch": 2.41,
1566
+ "learning_rate": 5.976829471578596e-05,
1567
+ "loss": 4.6155,
1568
+ "step": 4840
1569
+ },
1570
+ {
1571
+ "epoch": 2.42,
1572
+ "learning_rate": 5.960160026671112e-05,
1573
+ "loss": 4.6035,
1574
+ "step": 4860
1575
+ },
1576
+ {
1577
+ "epoch": 2.43,
1578
+ "learning_rate": 5.943490581763628e-05,
1579
+ "loss": 4.5941,
1580
+ "step": 4880
1581
+ },
1582
+ {
1583
+ "epoch": 2.44,
1584
+ "learning_rate": 5.926821136856143e-05,
1585
+ "loss": 4.6673,
1586
+ "step": 4900
1587
+ },
1588
+ {
1589
+ "epoch": 2.45,
1590
+ "learning_rate": 5.910151691948659e-05,
1591
+ "loss": 4.5854,
1592
+ "step": 4920
1593
+ },
1594
+ {
1595
+ "epoch": 2.46,
1596
+ "learning_rate": 5.8934822470411745e-05,
1597
+ "loss": 4.6004,
1598
+ "step": 4940
1599
+ },
1600
+ {
1601
+ "epoch": 2.47,
1602
+ "learning_rate": 5.876812802133689e-05,
1603
+ "loss": 4.5985,
1604
+ "step": 4960
1605
+ },
1606
+ {
1607
+ "epoch": 2.48,
1608
+ "learning_rate": 5.860143357226204e-05,
1609
+ "loss": 4.5952,
1610
+ "step": 4980
1611
+ },
1612
+ {
1613
+ "epoch": 2.49,
1614
+ "learning_rate": 5.84347391231872e-05,
1615
+ "loss": 4.574,
1616
+ "step": 5000
1617
+ },
1618
+ {
1619
+ "epoch": 2.5,
1620
+ "learning_rate": 5.828471411901983e-05,
1621
+ "loss": 4.549,
1622
+ "step": 5020
1623
+ },
1624
+ {
1625
+ "epoch": 2.51,
1626
+ "learning_rate": 5.811801966994499e-05,
1627
+ "loss": 4.6226,
1628
+ "step": 5040
1629
+ },
1630
+ {
1631
+ "epoch": 2.52,
1632
+ "learning_rate": 5.795132522087015e-05,
1633
+ "loss": 4.5922,
1634
+ "step": 5060
1635
+ },
1636
+ {
1637
+ "epoch": 2.53,
1638
+ "learning_rate": 5.77846307717953e-05,
1639
+ "loss": 4.5637,
1640
+ "step": 5080
1641
+ },
1642
+ {
1643
+ "epoch": 2.54,
1644
+ "learning_rate": 5.7617936322720455e-05,
1645
+ "loss": 4.5687,
1646
+ "step": 5100
1647
+ },
1648
+ {
1649
+ "epoch": 2.55,
1650
+ "learning_rate": 5.745124187364561e-05,
1651
+ "loss": 4.5703,
1652
+ "step": 5120
1653
+ },
1654
+ {
1655
+ "epoch": 2.56,
1656
+ "learning_rate": 5.7284547424570764e-05,
1657
+ "loss": 4.5802,
1658
+ "step": 5140
1659
+ },
1660
+ {
1661
+ "epoch": 2.57,
1662
+ "learning_rate": 5.711785297549592e-05,
1663
+ "loss": 4.5531,
1664
+ "step": 5160
1665
+ },
1666
+ {
1667
+ "epoch": 2.58,
1668
+ "learning_rate": 5.695115852642108e-05,
1669
+ "loss": 4.5561,
1670
+ "step": 5180
1671
+ },
1672
+ {
1673
+ "epoch": 2.59,
1674
+ "learning_rate": 5.6784464077346224e-05,
1675
+ "loss": 4.6081,
1676
+ "step": 5200
1677
+ },
1678
+ {
1679
+ "epoch": 2.59,
1680
+ "eval_validation_accuracy": 1.0,
1681
+ "eval_validation_loss": 5.37890625,
1682
+ "eval_validation_runtime": 338.279,
1683
+ "eval_validation_samples_per_second": 30.534,
1684
+ "eval_validation_steps_per_second": 1.274,
1685
+ "step": 5200
1686
+ },
1687
+ {
1688
+ "epoch": 2.6,
1689
+ "learning_rate": 5.6617769628271375e-05,
1690
+ "loss": 4.5354,
1691
+ "step": 5220
1692
+ },
1693
+ {
1694
+ "epoch": 2.61,
1695
+ "learning_rate": 5.645107517919653e-05,
1696
+ "loss": 4.5465,
1697
+ "step": 5240
1698
+ },
1699
+ {
1700
+ "epoch": 2.62,
1701
+ "learning_rate": 5.628438073012169e-05,
1702
+ "loss": 4.5041,
1703
+ "step": 5260
1704
+ },
1705
+ {
1706
+ "epoch": 2.63,
1707
+ "learning_rate": 5.611768628104684e-05,
1708
+ "loss": 4.5931,
1709
+ "step": 5280
1710
+ },
1711
+ {
1712
+ "epoch": 2.64,
1713
+ "learning_rate": 5.5950991831972e-05,
1714
+ "loss": 4.5416,
1715
+ "step": 5300
1716
+ },
1717
+ {
1718
+ "epoch": 2.65,
1719
+ "learning_rate": 5.578429738289716e-05,
1720
+ "loss": 4.5377,
1721
+ "step": 5320
1722
+ },
1723
+ {
1724
+ "epoch": 2.66,
1725
+ "learning_rate": 5.561760293382231e-05,
1726
+ "loss": 4.5618,
1727
+ "step": 5340
1728
+ },
1729
+ {
1730
+ "epoch": 2.67,
1731
+ "learning_rate": 5.5450908484747465e-05,
1732
+ "loss": 4.5547,
1733
+ "step": 5360
1734
+ },
1735
+ {
1736
+ "epoch": 2.68,
1737
+ "learning_rate": 5.528421403567261e-05,
1738
+ "loss": 4.556,
1739
+ "step": 5380
1740
+ },
1741
+ {
1742
+ "epoch": 2.69,
1743
+ "learning_rate": 5.511751958659777e-05,
1744
+ "loss": 4.5248,
1745
+ "step": 5400
1746
+ },
1747
+ {
1748
+ "epoch": 2.7,
1749
+ "learning_rate": 5.495082513752292e-05,
1750
+ "loss": 4.5663,
1751
+ "step": 5420
1752
+ },
1753
+ {
1754
+ "epoch": 2.71,
1755
+ "learning_rate": 5.4784130688448076e-05,
1756
+ "loss": 4.5785,
1757
+ "step": 5440
1758
+ },
1759
+ {
1760
+ "epoch": 2.72,
1761
+ "learning_rate": 5.4617436239373234e-05,
1762
+ "loss": 4.5752,
1763
+ "step": 5460
1764
+ },
1765
+ {
1766
+ "epoch": 2.73,
1767
+ "learning_rate": 5.4450741790298385e-05,
1768
+ "loss": 4.5014,
1769
+ "step": 5480
1770
+ },
1771
+ {
1772
+ "epoch": 2.74,
1773
+ "learning_rate": 5.428404734122354e-05,
1774
+ "loss": 4.5425,
1775
+ "step": 5500
1776
+ },
1777
+ {
1778
+ "epoch": 2.75,
1779
+ "learning_rate": 5.41173528921487e-05,
1780
+ "loss": 4.496,
1781
+ "step": 5520
1782
+ },
1783
+ {
1784
+ "epoch": 2.76,
1785
+ "learning_rate": 5.395065844307385e-05,
1786
+ "loss": 4.564,
1787
+ "step": 5540
1788
+ },
1789
+ {
1790
+ "epoch": 2.77,
1791
+ "learning_rate": 5.3783963993998995e-05,
1792
+ "loss": 4.4889,
1793
+ "step": 5560
1794
+ },
1795
+ {
1796
+ "epoch": 2.78,
1797
+ "learning_rate": 5.361726954492415e-05,
1798
+ "loss": 4.5218,
1799
+ "step": 5580
1800
+ },
1801
+ {
1802
+ "epoch": 2.79,
1803
+ "learning_rate": 5.345057509584931e-05,
1804
+ "loss": 4.5287,
1805
+ "step": 5600
1806
+ },
1807
+ {
1808
+ "epoch": 2.79,
1809
+ "eval_validation_accuracy": 1.0,
1810
+ "eval_validation_loss": 5.33203125,
1811
+ "eval_validation_runtime": 337.4194,
1812
+ "eval_validation_samples_per_second": 30.612,
1813
+ "eval_validation_steps_per_second": 1.277,
1814
+ "step": 5600
1815
+ },
1816
+ {
1817
+ "epoch": 2.8,
1818
+ "learning_rate": 5.328388064677446e-05,
1819
+ "loss": 4.5438,
1820
+ "step": 5620
1821
+ },
1822
+ {
1823
+ "epoch": 2.81,
1824
+ "learning_rate": 5.311718619769962e-05,
1825
+ "loss": 4.5329,
1826
+ "step": 5640
1827
+ },
1828
+ {
1829
+ "epoch": 2.82,
1830
+ "learning_rate": 5.295049174862478e-05,
1831
+ "loss": 4.5566,
1832
+ "step": 5660
1833
+ },
1834
+ {
1835
+ "epoch": 2.83,
1836
+ "learning_rate": 5.278379729954993e-05,
1837
+ "loss": 4.5243,
1838
+ "step": 5680
1839
+ },
1840
+ {
1841
+ "epoch": 2.84,
1842
+ "learning_rate": 5.2617102850475086e-05,
1843
+ "loss": 4.5396,
1844
+ "step": 5700
1845
+ },
1846
+ {
1847
+ "epoch": 2.85,
1848
+ "learning_rate": 5.245040840140023e-05,
1849
+ "loss": 4.5113,
1850
+ "step": 5720
1851
+ },
1852
+ {
1853
+ "epoch": 2.86,
1854
+ "learning_rate": 5.228371395232539e-05,
1855
+ "loss": 4.5003,
1856
+ "step": 5740
1857
+ },
1858
+ {
1859
+ "epoch": 2.87,
1860
+ "learning_rate": 5.211701950325054e-05,
1861
+ "loss": 4.546,
1862
+ "step": 5760
1863
+ },
1864
+ {
1865
+ "epoch": 2.88,
1866
+ "learning_rate": 5.19503250541757e-05,
1867
+ "loss": 4.5273,
1868
+ "step": 5780
1869
+ },
1870
+ {
1871
+ "epoch": 2.89,
1872
+ "learning_rate": 5.1783630605100854e-05,
1873
+ "loss": 4.4884,
1874
+ "step": 5800
1875
+ },
1876
+ {
1877
+ "epoch": 2.9,
1878
+ "learning_rate": 5.1616936156026005e-05,
1879
+ "loss": 4.5634,
1880
+ "step": 5820
1881
+ },
1882
+ {
1883
+ "epoch": 2.91,
1884
+ "learning_rate": 5.145024170695116e-05,
1885
+ "loss": 4.5258,
1886
+ "step": 5840
1887
+ },
1888
+ {
1889
+ "epoch": 2.92,
1890
+ "learning_rate": 5.128354725787632e-05,
1891
+ "loss": 4.5125,
1892
+ "step": 5860
1893
+ },
1894
+ {
1895
+ "epoch": 2.93,
1896
+ "learning_rate": 5.111685280880147e-05,
1897
+ "loss": 4.5234,
1898
+ "step": 5880
1899
+ },
1900
+ {
1901
+ "epoch": 2.94,
1902
+ "learning_rate": 5.0950158359726616e-05,
1903
+ "loss": 4.5286,
1904
+ "step": 5900
1905
+ },
1906
+ {
1907
+ "epoch": 2.95,
1908
+ "learning_rate": 5.0783463910651774e-05,
1909
+ "loss": 4.5492,
1910
+ "step": 5920
1911
+ },
1912
+ {
1913
+ "epoch": 2.96,
1914
+ "learning_rate": 5.061676946157693e-05,
1915
+ "loss": 4.5562,
1916
+ "step": 5940
1917
+ },
1918
+ {
1919
+ "epoch": 2.97,
1920
+ "learning_rate": 5.045007501250208e-05,
1921
+ "loss": 4.4468,
1922
+ "step": 5960
1923
+ },
1924
+ {
1925
+ "epoch": 2.98,
1926
+ "learning_rate": 5.028338056342724e-05,
1927
+ "loss": 4.4932,
1928
+ "step": 5980
1929
+ },
1930
+ {
1931
+ "epoch": 2.99,
1932
+ "learning_rate": 5.01166861143524e-05,
1933
+ "loss": 4.4833,
1934
+ "step": 6000
1935
+ },
1936
+ {
1937
+ "epoch": 2.99,
1938
+ "eval_validation_accuracy": 1.0,
1939
+ "eval_validation_loss": 5.29296875,
1940
+ "eval_validation_runtime": 338.8208,
1941
+ "eval_validation_samples_per_second": 30.485,
1942
+ "eval_validation_steps_per_second": 1.272,
1943
+ "step": 6000
1944
+ },
1945
+ {
1946
+ "epoch": 3.0,
1947
+ "learning_rate": 4.996666111018503e-05,
1948
+ "loss": 4.5214,
1949
+ "step": 6020
1950
+ },
1951
+ {
1952
+ "epoch": 3.01,
1953
+ "learning_rate": 4.979996666111019e-05,
1954
+ "loss": 4.4839,
1955
+ "step": 6040
1956
+ },
1957
+ {
1958
+ "epoch": 3.02,
1959
+ "learning_rate": 4.963327221203534e-05,
1960
+ "loss": 4.499,
1961
+ "step": 6060
1962
+ },
1963
+ {
1964
+ "epoch": 3.03,
1965
+ "learning_rate": 4.94665777629605e-05,
1966
+ "loss": 4.4424,
1967
+ "step": 6080
1968
+ },
1969
+ {
1970
+ "epoch": 3.04,
1971
+ "learning_rate": 4.929988331388565e-05,
1972
+ "loss": 4.4237,
1973
+ "step": 6100
1974
+ },
1975
+ {
1976
+ "epoch": 3.05,
1977
+ "learning_rate": 4.91331888648108e-05,
1978
+ "loss": 4.4712,
1979
+ "step": 6120
1980
+ },
1981
+ {
1982
+ "epoch": 3.06,
1983
+ "learning_rate": 4.896649441573596e-05,
1984
+ "loss": 4.4762,
1985
+ "step": 6140
1986
+ },
1987
+ {
1988
+ "epoch": 3.07,
1989
+ "learning_rate": 4.8799799966661115e-05,
1990
+ "loss": 4.4943,
1991
+ "step": 6160
1992
+ },
1993
+ {
1994
+ "epoch": 3.08,
1995
+ "learning_rate": 4.8633105517586266e-05,
1996
+ "loss": 4.4541,
1997
+ "step": 6180
1998
+ },
1999
+ {
2000
+ "epoch": 3.09,
2001
+ "learning_rate": 4.846641106851142e-05,
2002
+ "loss": 4.4592,
2003
+ "step": 6200
2004
+ },
2005
+ {
2006
+ "epoch": 3.1,
2007
+ "learning_rate": 4.8299716619436575e-05,
2008
+ "loss": 4.5027,
2009
+ "step": 6220
2010
+ },
2011
+ {
2012
+ "epoch": 3.11,
2013
+ "learning_rate": 4.813302217036173e-05,
2014
+ "loss": 4.4715,
2015
+ "step": 6240
2016
+ },
2017
+ {
2018
+ "epoch": 3.12,
2019
+ "learning_rate": 4.7966327721286884e-05,
2020
+ "loss": 4.4334,
2021
+ "step": 6260
2022
+ },
2023
+ {
2024
+ "epoch": 3.13,
2025
+ "learning_rate": 4.7799633272212034e-05,
2026
+ "loss": 4.455,
2027
+ "step": 6280
2028
+ },
2029
+ {
2030
+ "epoch": 3.14,
2031
+ "learning_rate": 4.763293882313719e-05,
2032
+ "loss": 4.4541,
2033
+ "step": 6300
2034
+ },
2035
+ {
2036
+ "epoch": 3.15,
2037
+ "learning_rate": 4.746624437406234e-05,
2038
+ "loss": 4.4482,
2039
+ "step": 6320
2040
+ },
2041
+ {
2042
+ "epoch": 3.16,
2043
+ "learning_rate": 4.72995499249875e-05,
2044
+ "loss": 4.4569,
2045
+ "step": 6340
2046
+ },
2047
+ {
2048
+ "epoch": 3.17,
2049
+ "learning_rate": 4.713285547591265e-05,
2050
+ "loss": 4.4142,
2051
+ "step": 6360
2052
+ },
2053
+ {
2054
+ "epoch": 3.18,
2055
+ "learning_rate": 4.696616102683781e-05,
2056
+ "loss": 4.4358,
2057
+ "step": 6380
2058
+ },
2059
+ {
2060
+ "epoch": 3.19,
2061
+ "learning_rate": 4.679946657776296e-05,
2062
+ "loss": 4.4256,
2063
+ "step": 6400
2064
+ },
2065
+ {
2066
+ "epoch": 3.19,
2067
+ "eval_validation_accuracy": 1.0,
2068
+ "eval_validation_loss": 5.26171875,
2069
+ "eval_validation_runtime": 340.5511,
2070
+ "eval_validation_samples_per_second": 30.33,
2071
+ "eval_validation_steps_per_second": 1.266,
2072
+ "step": 6400
2073
+ },
2074
+ {
2075
+ "epoch": 3.2,
2076
+ "learning_rate": 4.663277212868812e-05,
2077
+ "loss": 4.4243,
2078
+ "step": 6420
2079
+ },
2080
+ {
2081
+ "epoch": 3.21,
2082
+ "learning_rate": 4.6466077679613276e-05,
2083
+ "loss": 4.4286,
2084
+ "step": 6440
2085
+ },
2086
+ {
2087
+ "epoch": 3.22,
2088
+ "learning_rate": 4.629938323053842e-05,
2089
+ "loss": 4.3764,
2090
+ "step": 6460
2091
+ },
2092
+ {
2093
+ "epoch": 3.23,
2094
+ "learning_rate": 4.613268878146358e-05,
2095
+ "loss": 4.4376,
2096
+ "step": 6480
2097
+ },
2098
+ {
2099
+ "epoch": 3.24,
2100
+ "learning_rate": 4.5965994332388736e-05,
2101
+ "loss": 4.4027,
2102
+ "step": 6500
2103
+ },
2104
+ {
2105
+ "epoch": 3.25,
2106
+ "learning_rate": 4.579929988331389e-05,
2107
+ "loss": 4.4164,
2108
+ "step": 6520
2109
+ },
2110
+ {
2111
+ "epoch": 3.26,
2112
+ "learning_rate": 4.563260543423904e-05,
2113
+ "loss": 4.4505,
2114
+ "step": 6540
2115
+ },
2116
+ {
2117
+ "epoch": 3.27,
2118
+ "learning_rate": 4.5465910985164196e-05,
2119
+ "loss": 4.4574,
2120
+ "step": 6560
2121
+ },
2122
+ {
2123
+ "epoch": 3.28,
2124
+ "learning_rate": 4.529921653608935e-05,
2125
+ "loss": 4.4416,
2126
+ "step": 6580
2127
+ },
2128
+ {
2129
+ "epoch": 3.29,
2130
+ "learning_rate": 4.5132522087014504e-05,
2131
+ "loss": 4.4362,
2132
+ "step": 6600
2133
+ },
2134
+ {
2135
+ "epoch": 3.3,
2136
+ "learning_rate": 4.496582763793966e-05,
2137
+ "loss": 4.435,
2138
+ "step": 6620
2139
+ },
2140
+ {
2141
+ "epoch": 3.31,
2142
+ "learning_rate": 4.479913318886481e-05,
2143
+ "loss": 4.4101,
2144
+ "step": 6640
2145
+ },
2146
+ {
2147
+ "epoch": 3.32,
2148
+ "learning_rate": 4.4632438739789964e-05,
2149
+ "loss": 4.4322,
2150
+ "step": 6660
2151
+ },
2152
+ {
2153
+ "epoch": 3.33,
2154
+ "learning_rate": 4.446574429071512e-05,
2155
+ "loss": 4.4331,
2156
+ "step": 6680
2157
+ },
2158
+ {
2159
+ "epoch": 3.34,
2160
+ "learning_rate": 4.429904984164028e-05,
2161
+ "loss": 4.4413,
2162
+ "step": 6700
2163
+ },
2164
+ {
2165
+ "epoch": 3.35,
2166
+ "learning_rate": 4.413235539256543e-05,
2167
+ "loss": 4.4451,
2168
+ "step": 6720
2169
+ },
2170
+ {
2171
+ "epoch": 3.36,
2172
+ "learning_rate": 4.396566094349058e-05,
2173
+ "loss": 4.4744,
2174
+ "step": 6740
2175
+ },
2176
+ {
2177
+ "epoch": 3.37,
2178
+ "learning_rate": 4.379896649441574e-05,
2179
+ "loss": 4.4444,
2180
+ "step": 6760
2181
+ },
2182
+ {
2183
+ "epoch": 3.38,
2184
+ "learning_rate": 4.36322720453409e-05,
2185
+ "loss": 4.4582,
2186
+ "step": 6780
2187
+ },
2188
+ {
2189
+ "epoch": 3.39,
2190
+ "learning_rate": 4.346557759626604e-05,
2191
+ "loss": 4.4318,
2192
+ "step": 6800
2193
+ },
2194
+ {
2195
+ "epoch": 3.39,
2196
+ "eval_validation_accuracy": 1.0,
2197
+ "eval_validation_loss": 5.2265625,
2198
+ "eval_validation_runtime": 338.3936,
2199
+ "eval_validation_samples_per_second": 30.524,
2200
+ "eval_validation_steps_per_second": 1.274,
2201
+ "step": 6800
2202
+ },
2203
+ {
2204
+ "epoch": 3.4,
2205
+ "learning_rate": 4.32988831471912e-05,
2206
+ "loss": 4.4228,
2207
+ "step": 6820
2208
+ },
2209
+ {
2210
+ "epoch": 3.41,
2211
+ "learning_rate": 4.3132188698116357e-05,
2212
+ "loss": 4.4293,
2213
+ "step": 6840
2214
+ },
2215
+ {
2216
+ "epoch": 3.42,
2217
+ "learning_rate": 4.296549424904151e-05,
2218
+ "loss": 4.4289,
2219
+ "step": 6860
2220
+ },
2221
+ {
2222
+ "epoch": 3.43,
2223
+ "learning_rate": 4.2798799799966665e-05,
2224
+ "loss": 4.4076,
2225
+ "step": 6880
2226
+ },
2227
+ {
2228
+ "epoch": 3.44,
2229
+ "learning_rate": 4.2632105350891816e-05,
2230
+ "loss": 4.4333,
2231
+ "step": 6900
2232
+ },
2233
+ {
2234
+ "epoch": 3.45,
2235
+ "learning_rate": 4.2465410901816974e-05,
2236
+ "loss": 4.3814,
2237
+ "step": 6920
2238
+ },
2239
+ {
2240
+ "epoch": 3.46,
2241
+ "learning_rate": 4.2298716452742125e-05,
2242
+ "loss": 4.4256,
2243
+ "step": 6940
2244
+ },
2245
+ {
2246
+ "epoch": 3.47,
2247
+ "learning_rate": 4.213202200366728e-05,
2248
+ "loss": 4.4033,
2249
+ "step": 6960
2250
+ },
2251
+ {
2252
+ "epoch": 3.48,
2253
+ "learning_rate": 4.1965327554592434e-05,
2254
+ "loss": 4.4265,
2255
+ "step": 6980
2256
+ },
2257
+ {
2258
+ "epoch": 3.49,
2259
+ "learning_rate": 4.1798633105517585e-05,
2260
+ "loss": 4.3834,
2261
+ "step": 7000
2262
+ },
2263
+ {
2264
+ "epoch": 3.5,
2265
+ "learning_rate": 4.164860810135023e-05,
2266
+ "loss": 4.3796,
2267
+ "step": 7020
2268
+ },
2269
+ {
2270
+ "epoch": 3.51,
2271
+ "learning_rate": 4.148191365227538e-05,
2272
+ "loss": 4.4325,
2273
+ "step": 7040
2274
+ },
2275
+ {
2276
+ "epoch": 3.52,
2277
+ "learning_rate": 4.131521920320053e-05,
2278
+ "loss": 4.4101,
2279
+ "step": 7060
2280
+ },
2281
+ {
2282
+ "epoch": 3.53,
2283
+ "learning_rate": 4.114852475412569e-05,
2284
+ "loss": 4.4063,
2285
+ "step": 7080
2286
+ },
2287
+ {
2288
+ "epoch": 3.54,
2289
+ "learning_rate": 4.098183030505084e-05,
2290
+ "loss": 4.4233,
2291
+ "step": 7100
2292
+ },
2293
+ {
2294
+ "epoch": 3.55,
2295
+ "learning_rate": 4.0815135855976e-05,
2296
+ "loss": 4.3881,
2297
+ "step": 7120
2298
+ },
2299
+ {
2300
+ "epoch": 3.56,
2301
+ "learning_rate": 4.064844140690115e-05,
2302
+ "loss": 4.3984,
2303
+ "step": 7140
2304
+ },
2305
+ {
2306
+ "epoch": 3.57,
2307
+ "learning_rate": 4.048174695782631e-05,
2308
+ "loss": 4.3894,
2309
+ "step": 7160
2310
+ },
2311
+ {
2312
+ "epoch": 3.58,
2313
+ "learning_rate": 4.031505250875146e-05,
2314
+ "loss": 4.3899,
2315
+ "step": 7180
2316
+ },
2317
+ {
2318
+ "epoch": 3.59,
2319
+ "learning_rate": 4.014835805967662e-05,
2320
+ "loss": 4.4052,
2321
+ "step": 7200
2322
+ },
2323
+ {
2324
+ "epoch": 3.59,
2325
+ "eval_validation_accuracy": 1.0,
2326
+ "eval_validation_loss": 5.19921875,
2327
+ "eval_validation_runtime": 338.7907,
2328
+ "eval_validation_samples_per_second": 30.488,
2329
+ "eval_validation_steps_per_second": 1.272,
2330
+ "step": 7200
2331
+ },
2332
+ {
2333
+ "epoch": 3.6,
2334
+ "learning_rate": 3.998166361060177e-05,
2335
+ "loss": 4.3691,
2336
+ "step": 7220
2337
+ },
2338
+ {
2339
+ "epoch": 3.61,
2340
+ "learning_rate": 3.981496916152692e-05,
2341
+ "loss": 4.4268,
2342
+ "step": 7240
2343
+ },
2344
+ {
2345
+ "epoch": 3.62,
2346
+ "learning_rate": 3.964827471245208e-05,
2347
+ "loss": 4.4387,
2348
+ "step": 7260
2349
+ },
2350
+ {
2351
+ "epoch": 3.63,
2352
+ "learning_rate": 3.9481580263377235e-05,
2353
+ "loss": 4.4062,
2354
+ "step": 7280
2355
+ },
2356
+ {
2357
+ "epoch": 3.64,
2358
+ "learning_rate": 3.9314885814302386e-05,
2359
+ "loss": 4.4044,
2360
+ "step": 7300
2361
+ },
2362
+ {
2363
+ "epoch": 3.65,
2364
+ "learning_rate": 3.9148191365227537e-05,
2365
+ "loss": 4.4175,
2366
+ "step": 7320
2367
+ },
2368
+ {
2369
+ "epoch": 3.66,
2370
+ "learning_rate": 3.8981496916152694e-05,
2371
+ "loss": 4.421,
2372
+ "step": 7340
2373
+ },
2374
+ {
2375
+ "epoch": 3.67,
2376
+ "learning_rate": 3.881480246707785e-05,
2377
+ "loss": 4.3758,
2378
+ "step": 7360
2379
+ },
2380
+ {
2381
+ "epoch": 3.68,
2382
+ "learning_rate": 3.8648108018003e-05,
2383
+ "loss": 4.3597,
2384
+ "step": 7380
2385
+ },
2386
+ {
2387
+ "epoch": 3.69,
2388
+ "learning_rate": 3.8481413568928154e-05,
2389
+ "loss": 4.3891,
2390
+ "step": 7400
2391
+ },
2392
+ {
2393
+ "epoch": 3.7,
2394
+ "learning_rate": 3.831471911985331e-05,
2395
+ "loss": 4.3914,
2396
+ "step": 7420
2397
+ },
2398
+ {
2399
+ "epoch": 3.71,
2400
+ "learning_rate": 3.814802467077846e-05,
2401
+ "loss": 4.3861,
2402
+ "step": 7440
2403
+ },
2404
+ {
2405
+ "epoch": 3.72,
2406
+ "learning_rate": 3.798133022170362e-05,
2407
+ "loss": 4.3449,
2408
+ "step": 7460
2409
+ },
2410
+ {
2411
+ "epoch": 3.73,
2412
+ "learning_rate": 3.781463577262878e-05,
2413
+ "loss": 4.3784,
2414
+ "step": 7480
2415
+ },
2416
+ {
2417
+ "epoch": 3.74,
2418
+ "learning_rate": 3.764794132355393e-05,
2419
+ "loss": 4.4029,
2420
+ "step": 7500
2421
+ },
2422
+ {
2423
+ "epoch": 3.75,
2424
+ "learning_rate": 3.748124687447908e-05,
2425
+ "loss": 4.3866,
2426
+ "step": 7520
2427
+ },
2428
+ {
2429
+ "epoch": 3.76,
2430
+ "learning_rate": 3.731455242540424e-05,
2431
+ "loss": 4.4046,
2432
+ "step": 7540
2433
+ },
2434
+ {
2435
+ "epoch": 3.77,
2436
+ "learning_rate": 3.714785797632939e-05,
2437
+ "loss": 4.402,
2438
+ "step": 7560
2439
+ },
2440
+ {
2441
+ "epoch": 3.78,
2442
+ "learning_rate": 3.698116352725454e-05,
2443
+ "loss": 4.3447,
2444
+ "step": 7580
2445
+ },
2446
+ {
2447
+ "epoch": 3.79,
2448
+ "learning_rate": 3.68144690781797e-05,
2449
+ "loss": 4.4184,
2450
+ "step": 7600
2451
+ },
2452
+ {
2453
+ "epoch": 3.79,
2454
+ "eval_validation_accuracy": 1.0,
2455
+ "eval_validation_loss": 5.1796875,
2456
+ "eval_validation_runtime": 339.7216,
2457
+ "eval_validation_samples_per_second": 30.404,
2458
+ "eval_validation_steps_per_second": 1.269,
2459
+ "step": 7600
2460
+ },
2461
+ {
2462
+ "epoch": 3.8,
2463
+ "learning_rate": 3.6647774629104855e-05,
2464
+ "loss": 4.414,
2465
+ "step": 7620
2466
+ },
2467
+ {
2468
+ "epoch": 3.81,
2469
+ "learning_rate": 3.6481080180030006e-05,
2470
+ "loss": 4.4022,
2471
+ "step": 7640
2472
+ },
2473
+ {
2474
+ "epoch": 3.82,
2475
+ "learning_rate": 3.631438573095516e-05,
2476
+ "loss": 4.3949,
2477
+ "step": 7660
2478
+ },
2479
+ {
2480
+ "epoch": 3.83,
2481
+ "learning_rate": 3.6147691281880315e-05,
2482
+ "loss": 4.3867,
2483
+ "step": 7680
2484
+ },
2485
+ {
2486
+ "epoch": 3.84,
2487
+ "learning_rate": 3.5980996832805466e-05,
2488
+ "loss": 4.3992,
2489
+ "step": 7700
2490
+ },
2491
+ {
2492
+ "epoch": 3.85,
2493
+ "learning_rate": 3.5814302383730624e-05,
2494
+ "loss": 4.3692,
2495
+ "step": 7720
2496
+ },
2497
+ {
2498
+ "epoch": 3.86,
2499
+ "learning_rate": 3.564760793465578e-05,
2500
+ "loss": 4.3693,
2501
+ "step": 7740
2502
+ },
2503
+ {
2504
+ "epoch": 3.87,
2505
+ "learning_rate": 3.548091348558093e-05,
2506
+ "loss": 4.3892,
2507
+ "step": 7760
2508
+ },
2509
+ {
2510
+ "epoch": 3.88,
2511
+ "learning_rate": 3.531421903650608e-05,
2512
+ "loss": 4.3402,
2513
+ "step": 7780
2514
+ },
2515
+ {
2516
+ "epoch": 3.89,
2517
+ "learning_rate": 3.514752458743124e-05,
2518
+ "loss": 4.3698,
2519
+ "step": 7800
2520
+ },
2521
+ {
2522
+ "epoch": 3.9,
2523
+ "learning_rate": 3.49808301383564e-05,
2524
+ "loss": 4.4092,
2525
+ "step": 7820
2526
+ },
2527
+ {
2528
+ "epoch": 3.91,
2529
+ "learning_rate": 3.481413568928154e-05,
2530
+ "loss": 4.3575,
2531
+ "step": 7840
2532
+ },
2533
+ {
2534
+ "epoch": 3.92,
2535
+ "learning_rate": 3.46474412402067e-05,
2536
+ "loss": 4.3515,
2537
+ "step": 7860
2538
+ },
2539
+ {
2540
+ "epoch": 3.93,
2541
+ "learning_rate": 3.448074679113186e-05,
2542
+ "loss": 4.4052,
2543
+ "step": 7880
2544
+ },
2545
+ {
2546
+ "epoch": 3.94,
2547
+ "learning_rate": 3.431405234205701e-05,
2548
+ "loss": 4.3975,
2549
+ "step": 7900
2550
+ },
2551
+ {
2552
+ "epoch": 3.95,
2553
+ "learning_rate": 3.414735789298217e-05,
2554
+ "loss": 4.4063,
2555
+ "step": 7920
2556
+ },
2557
+ {
2558
+ "epoch": 3.96,
2559
+ "learning_rate": 3.398066344390732e-05,
2560
+ "loss": 4.3948,
2561
+ "step": 7940
2562
+ },
2563
+ {
2564
+ "epoch": 3.97,
2565
+ "learning_rate": 3.3813968994832476e-05,
2566
+ "loss": 4.3733,
2567
+ "step": 7960
2568
+ },
2569
+ {
2570
+ "epoch": 3.98,
2571
+ "learning_rate": 3.364727454575763e-05,
2572
+ "loss": 4.3302,
2573
+ "step": 7980
2574
+ },
2575
+ {
2576
+ "epoch": 3.99,
2577
+ "learning_rate": 3.3480580096682785e-05,
2578
+ "loss": 4.3676,
2579
+ "step": 8000
2580
+ },
2581
+ {
2582
+ "epoch": 3.99,
2583
+ "eval_validation_accuracy": 1.0,
2584
+ "eval_validation_loss": 5.15625,
2585
+ "eval_validation_runtime": 338.5475,
2586
+ "eval_validation_samples_per_second": 30.51,
2587
+ "eval_validation_steps_per_second": 1.273,
2588
+ "step": 8000
2589
+ },
2590
+ {
2591
+ "epoch": 4.0,
2592
+ "learning_rate": 3.333055509251542e-05,
2593
+ "loss": 4.4045,
2594
+ "step": 8020
2595
+ },
2596
+ {
2597
+ "epoch": 4.01,
2598
+ "learning_rate": 3.3163860643440576e-05,
2599
+ "loss": 4.3992,
2600
+ "step": 8040
2601
+ },
2602
+ {
2603
+ "epoch": 4.02,
2604
+ "learning_rate": 3.299716619436573e-05,
2605
+ "loss": 4.3823,
2606
+ "step": 8060
2607
+ },
2608
+ {
2609
+ "epoch": 4.03,
2610
+ "learning_rate": 3.2830471745290884e-05,
2611
+ "loss": 4.3124,
2612
+ "step": 8080
2613
+ },
2614
+ {
2615
+ "epoch": 4.04,
2616
+ "learning_rate": 3.2663777296216035e-05,
2617
+ "loss": 4.3796,
2618
+ "step": 8100
2619
+ },
2620
+ {
2621
+ "epoch": 4.05,
2622
+ "learning_rate": 3.249708284714119e-05,
2623
+ "loss": 4.3279,
2624
+ "step": 8120
2625
+ },
2626
+ {
2627
+ "epoch": 4.06,
2628
+ "learning_rate": 3.2330388398066344e-05,
2629
+ "loss": 4.3688,
2630
+ "step": 8140
2631
+ },
2632
+ {
2633
+ "epoch": 4.07,
2634
+ "learning_rate": 3.21636939489915e-05,
2635
+ "loss": 4.3267,
2636
+ "step": 8160
2637
+ },
2638
+ {
2639
+ "epoch": 4.08,
2640
+ "learning_rate": 3.199699949991665e-05,
2641
+ "loss": 4.3737,
2642
+ "step": 8180
2643
+ },
2644
+ {
2645
+ "epoch": 4.09,
2646
+ "learning_rate": 3.183030505084181e-05,
2647
+ "loss": 4.3382,
2648
+ "step": 8200
2649
+ },
2650
+ {
2651
+ "epoch": 4.1,
2652
+ "learning_rate": 3.166361060176696e-05,
2653
+ "loss": 4.3581,
2654
+ "step": 8220
2655
+ },
2656
+ {
2657
+ "epoch": 4.11,
2658
+ "learning_rate": 3.149691615269212e-05,
2659
+ "loss": 4.3538,
2660
+ "step": 8240
2661
+ },
2662
+ {
2663
+ "epoch": 4.12,
2664
+ "learning_rate": 3.133022170361727e-05,
2665
+ "loss": 4.3174,
2666
+ "step": 8260
2667
+ },
2668
+ {
2669
+ "epoch": 4.13,
2670
+ "learning_rate": 3.116352725454242e-05,
2671
+ "loss": 4.3892,
2672
+ "step": 8280
2673
+ },
2674
+ {
2675
+ "epoch": 4.14,
2676
+ "learning_rate": 3.099683280546758e-05,
2677
+ "loss": 4.3253,
2678
+ "step": 8300
2679
+ },
2680
+ {
2681
+ "epoch": 4.15,
2682
+ "learning_rate": 3.0830138356392737e-05,
2683
+ "loss": 4.3314,
2684
+ "step": 8320
2685
+ },
2686
+ {
2687
+ "epoch": 4.16,
2688
+ "learning_rate": 3.066344390731789e-05,
2689
+ "loss": 4.3372,
2690
+ "step": 8340
2691
+ },
2692
+ {
2693
+ "epoch": 4.17,
2694
+ "learning_rate": 3.049674945824304e-05,
2695
+ "loss": 4.3271,
2696
+ "step": 8360
2697
+ },
2698
+ {
2699
+ "epoch": 4.18,
2700
+ "learning_rate": 3.0330055009168196e-05,
2701
+ "loss": 4.3183,
2702
+ "step": 8380
2703
+ },
2704
+ {
2705
+ "epoch": 4.19,
2706
+ "learning_rate": 3.016336056009335e-05,
2707
+ "loss": 4.3448,
2708
+ "step": 8400
2709
+ },
2710
+ {
2711
+ "epoch": 4.19,
2712
+ "eval_validation_accuracy": 1.0,
2713
+ "eval_validation_loss": 5.140625,
2714
+ "eval_validation_runtime": 338.033,
2715
+ "eval_validation_samples_per_second": 30.556,
2716
+ "eval_validation_steps_per_second": 1.275,
2717
+ "step": 8400
2718
+ },
2719
+ {
2720
+ "epoch": 4.2,
2721
+ "learning_rate": 2.9996666111018505e-05,
2722
+ "loss": 4.347,
2723
+ "step": 8420
2724
+ },
2725
+ {
2726
+ "epoch": 4.21,
2727
+ "learning_rate": 2.9829971661943656e-05,
2728
+ "loss": 4.2873,
2729
+ "step": 8440
2730
+ },
2731
+ {
2732
+ "epoch": 4.22,
2733
+ "learning_rate": 2.966327721286881e-05,
2734
+ "loss": 4.3227,
2735
+ "step": 8460
2736
+ },
2737
+ {
2738
+ "epoch": 4.23,
2739
+ "learning_rate": 2.9496582763793968e-05,
2740
+ "loss": 4.3491,
2741
+ "step": 8480
2742
+ },
2743
+ {
2744
+ "epoch": 4.24,
2745
+ "learning_rate": 2.9329888314719122e-05,
2746
+ "loss": 4.3198,
2747
+ "step": 8500
2748
+ },
2749
+ {
2750
+ "epoch": 4.25,
2751
+ "learning_rate": 2.9163193865644273e-05,
2752
+ "loss": 4.3333,
2753
+ "step": 8520
2754
+ },
2755
+ {
2756
+ "epoch": 4.26,
2757
+ "learning_rate": 2.8996499416569428e-05,
2758
+ "loss": 4.3269,
2759
+ "step": 8540
2760
+ },
2761
+ {
2762
+ "epoch": 4.27,
2763
+ "learning_rate": 2.8829804967494582e-05,
2764
+ "loss": 4.3557,
2765
+ "step": 8560
2766
+ },
2767
+ {
2768
+ "epoch": 4.28,
2769
+ "learning_rate": 2.866311051841974e-05,
2770
+ "loss": 4.336,
2771
+ "step": 8580
2772
+ },
2773
+ {
2774
+ "epoch": 4.29,
2775
+ "learning_rate": 2.8496416069344894e-05,
2776
+ "loss": 4.325,
2777
+ "step": 8600
2778
+ },
2779
+ {
2780
+ "epoch": 4.3,
2781
+ "learning_rate": 2.8329721620270045e-05,
2782
+ "loss": 4.2758,
2783
+ "step": 8620
2784
+ },
2785
+ {
2786
+ "epoch": 4.31,
2787
+ "learning_rate": 2.81630271711952e-05,
2788
+ "loss": 4.3227,
2789
+ "step": 8640
2790
+ },
2791
+ {
2792
+ "epoch": 4.32,
2793
+ "learning_rate": 2.7996332722120354e-05,
2794
+ "loss": 4.373,
2795
+ "step": 8660
2796
+ },
2797
+ {
2798
+ "epoch": 4.33,
2799
+ "learning_rate": 2.782963827304551e-05,
2800
+ "loss": 4.3084,
2801
+ "step": 8680
2802
+ },
2803
+ {
2804
+ "epoch": 4.34,
2805
+ "learning_rate": 2.766294382397066e-05,
2806
+ "loss": 4.364,
2807
+ "step": 8700
2808
+ },
2809
+ {
2810
+ "epoch": 4.35,
2811
+ "learning_rate": 2.7496249374895817e-05,
2812
+ "loss": 4.3126,
2813
+ "step": 8720
2814
+ },
2815
+ {
2816
+ "epoch": 4.36,
2817
+ "learning_rate": 2.732955492582097e-05,
2818
+ "loss": 4.3171,
2819
+ "step": 8740
2820
+ },
2821
+ {
2822
+ "epoch": 4.37,
2823
+ "learning_rate": 2.7162860476746126e-05,
2824
+ "loss": 4.3314,
2825
+ "step": 8760
2826
+ },
2827
+ {
2828
+ "epoch": 4.38,
2829
+ "learning_rate": 2.6996166027671283e-05,
2830
+ "loss": 4.3669,
2831
+ "step": 8780
2832
+ },
2833
+ {
2834
+ "epoch": 4.39,
2835
+ "learning_rate": 2.682947157859643e-05,
2836
+ "loss": 4.3437,
2837
+ "step": 8800
2838
+ },
2839
+ {
2840
+ "epoch": 4.39,
2841
+ "eval_validation_accuracy": 1.0,
2842
+ "eval_validation_loss": 5.12109375,
2843
+ "eval_validation_runtime": 339.6825,
2844
+ "eval_validation_samples_per_second": 30.408,
2845
+ "eval_validation_steps_per_second": 1.269,
2846
+ "step": 8800
2847
+ },
2848
+ {
2849
+ "epoch": 4.4,
2850
+ "learning_rate": 2.666277712952159e-05,
2851
+ "loss": 4.3182,
2852
+ "step": 8820
2853
+ },
2854
+ {
2855
+ "epoch": 4.41,
2856
+ "learning_rate": 2.6496082680446743e-05,
2857
+ "loss": 4.3251,
2858
+ "step": 8840
2859
+ },
2860
+ {
2861
+ "epoch": 4.42,
2862
+ "learning_rate": 2.6329388231371897e-05,
2863
+ "loss": 4.3391,
2864
+ "step": 8860
2865
+ },
2866
+ {
2867
+ "epoch": 4.43,
2868
+ "learning_rate": 2.616269378229705e-05,
2869
+ "loss": 4.3208,
2870
+ "step": 8880
2871
+ },
2872
+ {
2873
+ "epoch": 4.44,
2874
+ "learning_rate": 2.5995999333222203e-05,
2875
+ "loss": 4.321,
2876
+ "step": 8900
2877
+ },
2878
+ {
2879
+ "epoch": 4.45,
2880
+ "learning_rate": 2.582930488414736e-05,
2881
+ "loss": 4.3163,
2882
+ "step": 8920
2883
+ },
2884
+ {
2885
+ "epoch": 4.46,
2886
+ "learning_rate": 2.5662610435072515e-05,
2887
+ "loss": 4.2582,
2888
+ "step": 8940
2889
+ },
2890
+ {
2891
+ "epoch": 4.47,
2892
+ "learning_rate": 2.5495915985997666e-05,
2893
+ "loss": 4.3347,
2894
+ "step": 8960
2895
+ },
2896
+ {
2897
+ "epoch": 4.48,
2898
+ "learning_rate": 2.532922153692282e-05,
2899
+ "loss": 4.2907,
2900
+ "step": 8980
2901
+ },
2902
+ {
2903
+ "epoch": 4.49,
2904
+ "learning_rate": 2.5162527087847975e-05,
2905
+ "loss": 4.3013,
2906
+ "step": 9000
2907
+ },
2908
+ {
2909
+ "epoch": 4.5,
2910
+ "learning_rate": 2.5012502083680618e-05,
2911
+ "loss": 4.2908,
2912
+ "step": 9020
2913
+ },
2914
+ {
2915
+ "epoch": 4.51,
2916
+ "learning_rate": 2.484580763460577e-05,
2917
+ "loss": 4.341,
2918
+ "step": 9040
2919
+ },
2920
+ {
2921
+ "epoch": 4.52,
2922
+ "learning_rate": 2.4679113185530923e-05,
2923
+ "loss": 4.2858,
2924
+ "step": 9060
2925
+ },
2926
+ {
2927
+ "epoch": 4.53,
2928
+ "learning_rate": 2.4512418736456078e-05,
2929
+ "loss": 4.3251,
2930
+ "step": 9080
2931
+ },
2932
+ {
2933
+ "epoch": 4.54,
2934
+ "learning_rate": 2.434572428738123e-05,
2935
+ "loss": 4.334,
2936
+ "step": 9100
2937
+ },
2938
+ {
2939
+ "epoch": 4.55,
2940
+ "learning_rate": 2.4179029838306386e-05,
2941
+ "loss": 4.3635,
2942
+ "step": 9120
2943
+ },
2944
+ {
2945
+ "epoch": 4.56,
2946
+ "learning_rate": 2.4012335389231537e-05,
2947
+ "loss": 4.3153,
2948
+ "step": 9140
2949
+ },
2950
+ {
2951
+ "epoch": 4.57,
2952
+ "learning_rate": 2.3845640940156695e-05,
2953
+ "loss": 4.2986,
2954
+ "step": 9160
2955
+ },
2956
+ {
2957
+ "epoch": 4.58,
2958
+ "learning_rate": 2.367894649108185e-05,
2959
+ "loss": 4.3384,
2960
+ "step": 9180
2961
+ },
2962
+ {
2963
+ "epoch": 4.59,
2964
+ "learning_rate": 2.3512252042007e-05,
2965
+ "loss": 4.2735,
2966
+ "step": 9200
2967
+ },
2968
+ {
2969
+ "epoch": 4.59,
2970
+ "eval_validation_accuracy": 1.0,
2971
+ "eval_validation_loss": 5.10546875,
2972
+ "eval_validation_runtime": 338.1676,
2973
+ "eval_validation_samples_per_second": 30.544,
2974
+ "eval_validation_steps_per_second": 1.275,
2975
+ "step": 9200
2976
+ },
2977
+ {
2978
+ "epoch": 4.6,
2979
+ "learning_rate": 2.3345557592932158e-05,
2980
+ "loss": 4.2789,
2981
+ "step": 9220
2982
+ },
2983
+ {
2984
+ "epoch": 4.61,
2985
+ "learning_rate": 2.317886314385731e-05,
2986
+ "loss": 4.3535,
2987
+ "step": 9240
2988
+ },
2989
+ {
2990
+ "epoch": 4.62,
2991
+ "learning_rate": 2.3012168694782467e-05,
2992
+ "loss": 4.347,
2993
+ "step": 9260
2994
+ },
2995
+ {
2996
+ "epoch": 4.63,
2997
+ "learning_rate": 2.2845474245707618e-05,
2998
+ "loss": 4.2697,
2999
+ "step": 9280
3000
+ },
3001
+ {
3002
+ "epoch": 4.64,
3003
+ "learning_rate": 2.2678779796632772e-05,
3004
+ "loss": 4.3388,
3005
+ "step": 9300
3006
+ },
3007
+ {
3008
+ "epoch": 4.65,
3009
+ "learning_rate": 2.2512085347557927e-05,
3010
+ "loss": 4.3372,
3011
+ "step": 9320
3012
+ },
3013
+ {
3014
+ "epoch": 4.66,
3015
+ "learning_rate": 2.234539089848308e-05,
3016
+ "loss": 4.3309,
3017
+ "step": 9340
3018
+ },
3019
+ {
3020
+ "epoch": 4.67,
3021
+ "learning_rate": 2.2178696449408235e-05,
3022
+ "loss": 4.2927,
3023
+ "step": 9360
3024
+ },
3025
+ {
3026
+ "epoch": 4.68,
3027
+ "learning_rate": 2.201200200033339e-05,
3028
+ "loss": 4.2814,
3029
+ "step": 9380
3030
+ },
3031
+ {
3032
+ "epoch": 4.69,
3033
+ "learning_rate": 2.1845307551258544e-05,
3034
+ "loss": 4.3071,
3035
+ "step": 9400
3036
+ },
3037
+ {
3038
+ "epoch": 4.7,
3039
+ "learning_rate": 2.1678613102183698e-05,
3040
+ "loss": 4.2728,
3041
+ "step": 9420
3042
+ },
3043
+ {
3044
+ "epoch": 4.71,
3045
+ "learning_rate": 2.1511918653108853e-05,
3046
+ "loss": 4.3184,
3047
+ "step": 9440
3048
+ },
3049
+ {
3050
+ "epoch": 4.72,
3051
+ "learning_rate": 2.1345224204034007e-05,
3052
+ "loss": 4.2919,
3053
+ "step": 9460
3054
+ },
3055
+ {
3056
+ "epoch": 4.73,
3057
+ "learning_rate": 2.117852975495916e-05,
3058
+ "loss": 4.293,
3059
+ "step": 9480
3060
+ },
3061
+ {
3062
+ "epoch": 4.74,
3063
+ "learning_rate": 2.1011835305884316e-05,
3064
+ "loss": 4.3149,
3065
+ "step": 9500
3066
+ },
3067
+ {
3068
+ "epoch": 4.75,
3069
+ "learning_rate": 2.084514085680947e-05,
3070
+ "loss": 4.2432,
3071
+ "step": 9520
3072
+ },
3073
+ {
3074
+ "epoch": 4.76,
3075
+ "learning_rate": 2.067844640773462e-05,
3076
+ "loss": 4.2945,
3077
+ "step": 9540
3078
+ },
3079
+ {
3080
+ "epoch": 4.77,
3081
+ "learning_rate": 2.051175195865978e-05,
3082
+ "loss": 4.2773,
3083
+ "step": 9560
3084
+ },
3085
+ {
3086
+ "epoch": 4.78,
3087
+ "learning_rate": 2.034505750958493e-05,
3088
+ "loss": 4.3129,
3089
+ "step": 9580
3090
+ },
3091
+ {
3092
+ "epoch": 4.79,
3093
+ "learning_rate": 2.0178363060510088e-05,
3094
+ "loss": 4.3002,
3095
+ "step": 9600
3096
+ },
3097
+ {
3098
+ "epoch": 4.79,
3099
+ "eval_validation_accuracy": 1.0,
3100
+ "eval_validation_loss": 5.08984375,
3101
+ "eval_validation_runtime": 339.2017,
3102
+ "eval_validation_samples_per_second": 30.451,
3103
+ "eval_validation_steps_per_second": 1.271,
3104
+ "step": 9600
3105
+ },
3106
+ {
3107
+ "epoch": 4.8,
3108
+ "learning_rate": 2.0011668611435242e-05,
3109
+ "loss": 4.3626,
3110
+ "step": 9620
3111
+ },
3112
+ {
3113
+ "epoch": 4.81,
3114
+ "learning_rate": 1.9844974162360393e-05,
3115
+ "loss": 4.3427,
3116
+ "step": 9640
3117
+ },
3118
+ {
3119
+ "epoch": 4.82,
3120
+ "learning_rate": 1.967827971328555e-05,
3121
+ "loss": 4.301,
3122
+ "step": 9660
3123
+ },
3124
+ {
3125
+ "epoch": 4.83,
3126
+ "learning_rate": 1.95115852642107e-05,
3127
+ "loss": 4.3136,
3128
+ "step": 9680
3129
+ },
3130
+ {
3131
+ "epoch": 4.84,
3132
+ "learning_rate": 1.934489081513586e-05,
3133
+ "loss": 4.3024,
3134
+ "step": 9700
3135
+ },
3136
+ {
3137
+ "epoch": 4.85,
3138
+ "learning_rate": 1.917819636606101e-05,
3139
+ "loss": 4.2682,
3140
+ "step": 9720
3141
+ },
3142
+ {
3143
+ "epoch": 4.86,
3144
+ "learning_rate": 1.9011501916986165e-05,
3145
+ "loss": 4.2849,
3146
+ "step": 9740
3147
+ },
3148
+ {
3149
+ "epoch": 4.87,
3150
+ "learning_rate": 1.884480746791132e-05,
3151
+ "loss": 4.306,
3152
+ "step": 9760
3153
+ },
3154
+ {
3155
+ "epoch": 4.88,
3156
+ "learning_rate": 1.8678113018836473e-05,
3157
+ "loss": 4.2644,
3158
+ "step": 9780
3159
+ },
3160
+ {
3161
+ "epoch": 4.89,
3162
+ "learning_rate": 1.8511418569761628e-05,
3163
+ "loss": 4.3479,
3164
+ "step": 9800
3165
+ },
3166
+ {
3167
+ "epoch": 4.9,
3168
+ "learning_rate": 1.8344724120686782e-05,
3169
+ "loss": 4.3379,
3170
+ "step": 9820
3171
+ },
3172
+ {
3173
+ "epoch": 4.91,
3174
+ "learning_rate": 1.8178029671611936e-05,
3175
+ "loss": 4.2851,
3176
+ "step": 9840
3177
+ },
3178
+ {
3179
+ "epoch": 4.92,
3180
+ "learning_rate": 1.801133522253709e-05,
3181
+ "loss": 4.2967,
3182
+ "step": 9860
3183
+ },
3184
+ {
3185
+ "epoch": 4.93,
3186
+ "learning_rate": 1.7844640773462245e-05,
3187
+ "loss": 4.2853,
3188
+ "step": 9880
3189
+ },
3190
+ {
3191
+ "epoch": 4.94,
3192
+ "learning_rate": 1.76779463243874e-05,
3193
+ "loss": 4.2708,
3194
+ "step": 9900
3195
+ },
3196
+ {
3197
+ "epoch": 4.95,
3198
+ "learning_rate": 1.7511251875312554e-05,
3199
+ "loss": 4.3077,
3200
+ "step": 9920
3201
+ },
3202
+ {
3203
+ "epoch": 4.96,
3204
+ "learning_rate": 1.7344557426237705e-05,
3205
+ "loss": 4.2787,
3206
+ "step": 9940
3207
+ },
3208
+ {
3209
+ "epoch": 4.97,
3210
+ "learning_rate": 1.7177862977162863e-05,
3211
+ "loss": 4.3246,
3212
+ "step": 9960
3213
+ },
3214
+ {
3215
+ "epoch": 4.98,
3216
+ "learning_rate": 1.7011168528088014e-05,
3217
+ "loss": 4.2724,
3218
+ "step": 9980
3219
+ },
3220
+ {
3221
+ "epoch": 4.99,
3222
+ "learning_rate": 1.684447407901317e-05,
3223
+ "loss": 4.2508,
3224
+ "step": 10000
3225
+ },
3226
+ {
3227
+ "epoch": 4.99,
3228
+ "eval_validation_accuracy": 1.0,
3229
+ "eval_validation_loss": 5.078125,
3230
+ "eval_validation_runtime": 338.4305,
3231
+ "eval_validation_samples_per_second": 30.52,
3232
+ "eval_validation_steps_per_second": 1.274,
3233
+ "step": 10000
3234
+ },
3235
+ {
3236
+ "epoch": 5.0,
3237
+ "learning_rate": 1.6677779629938322e-05,
3238
+ "loss": 4.2658,
3239
+ "step": 10020
3240
+ },
3241
+ {
3242
+ "epoch": 5.0,
3243
+ "learning_rate": 1.6527754625770962e-05,
3244
+ "loss": 4.3492,
3245
+ "step": 10040
3246
+ },
3247
+ {
3248
+ "epoch": 5.01,
3249
+ "learning_rate": 1.6361060176696117e-05,
3250
+ "loss": 4.3095,
3251
+ "step": 10060
3252
+ },
3253
+ {
3254
+ "epoch": 5.02,
3255
+ "learning_rate": 1.619436572762127e-05,
3256
+ "loss": 4.2266,
3257
+ "step": 10080
3258
+ },
3259
+ {
3260
+ "epoch": 5.03,
3261
+ "learning_rate": 1.6027671278546425e-05,
3262
+ "loss": 4.2572,
3263
+ "step": 10100
3264
+ },
3265
+ {
3266
+ "epoch": 5.04,
3267
+ "learning_rate": 1.586097682947158e-05,
3268
+ "loss": 4.2703,
3269
+ "step": 10120
3270
+ },
3271
+ {
3272
+ "epoch": 5.05,
3273
+ "learning_rate": 1.5694282380396734e-05,
3274
+ "loss": 4.227,
3275
+ "step": 10140
3276
+ },
3277
+ {
3278
+ "epoch": 5.06,
3279
+ "learning_rate": 1.552758793132189e-05,
3280
+ "loss": 4.2915,
3281
+ "step": 10160
3282
+ },
3283
+ {
3284
+ "epoch": 5.07,
3285
+ "learning_rate": 1.536089348224704e-05,
3286
+ "loss": 4.3084,
3287
+ "step": 10180
3288
+ },
3289
+ {
3290
+ "epoch": 5.08,
3291
+ "learning_rate": 1.5194199033172195e-05,
3292
+ "loss": 4.235,
3293
+ "step": 10200
3294
+ },
3295
+ {
3296
+ "epoch": 5.09,
3297
+ "learning_rate": 1.5027504584097351e-05,
3298
+ "loss": 4.3103,
3299
+ "step": 10220
3300
+ },
3301
+ {
3302
+ "epoch": 5.1,
3303
+ "learning_rate": 1.4860810135022504e-05,
3304
+ "loss": 4.2972,
3305
+ "step": 10240
3306
+ },
3307
+ {
3308
+ "epoch": 5.11,
3309
+ "learning_rate": 1.469411568594766e-05,
3310
+ "loss": 4.2973,
3311
+ "step": 10260
3312
+ },
3313
+ {
3314
+ "epoch": 5.12,
3315
+ "learning_rate": 1.4527421236872813e-05,
3316
+ "loss": 4.2718,
3317
+ "step": 10280
3318
+ },
3319
+ {
3320
+ "epoch": 5.13,
3321
+ "learning_rate": 1.4360726787797967e-05,
3322
+ "loss": 4.2749,
3323
+ "step": 10300
3324
+ },
3325
+ {
3326
+ "epoch": 5.14,
3327
+ "learning_rate": 1.419403233872312e-05,
3328
+ "loss": 4.2932,
3329
+ "step": 10320
3330
+ },
3331
+ {
3332
+ "epoch": 5.15,
3333
+ "learning_rate": 1.4027337889648276e-05,
3334
+ "loss": 4.2442,
3335
+ "step": 10340
3336
+ },
3337
+ {
3338
+ "epoch": 5.16,
3339
+ "learning_rate": 1.3860643440573429e-05,
3340
+ "loss": 4.2829,
3341
+ "step": 10360
3342
+ },
3343
+ {
3344
+ "epoch": 5.17,
3345
+ "learning_rate": 1.3693948991498585e-05,
3346
+ "loss": 4.2797,
3347
+ "step": 10380
3348
+ },
3349
+ {
3350
+ "epoch": 5.18,
3351
+ "learning_rate": 1.3527254542423737e-05,
3352
+ "loss": 4.276,
3353
+ "step": 10400
3354
+ },
3355
+ {
3356
+ "epoch": 5.18,
3357
+ "eval_validation_accuracy": 1.0,
3358
+ "eval_validation_loss": 5.07421875,
3359
+ "eval_validation_runtime": 339.3821,
3360
+ "eval_validation_samples_per_second": 30.435,
3361
+ "eval_validation_steps_per_second": 1.27,
3362
+ "step": 10400
3363
+ },
3364
+ {
3365
+ "epoch": 5.19,
3366
+ "learning_rate": 1.3360560093348892e-05,
3367
+ "loss": 4.3034,
3368
+ "step": 10420
3369
+ },
3370
+ {
3371
+ "epoch": 5.2,
3372
+ "learning_rate": 1.3193865644274044e-05,
3373
+ "loss": 4.24,
3374
+ "step": 10440
3375
+ },
3376
+ {
3377
+ "epoch": 5.21,
3378
+ "learning_rate": 1.30271711951992e-05,
3379
+ "loss": 4.291,
3380
+ "step": 10460
3381
+ },
3382
+ {
3383
+ "epoch": 5.22,
3384
+ "learning_rate": 1.2860476746124356e-05,
3385
+ "loss": 4.2267,
3386
+ "step": 10480
3387
+ },
3388
+ {
3389
+ "epoch": 5.23,
3390
+ "learning_rate": 1.2693782297049509e-05,
3391
+ "loss": 4.2393,
3392
+ "step": 10500
3393
+ },
3394
+ {
3395
+ "epoch": 5.24,
3396
+ "learning_rate": 1.2527087847974663e-05,
3397
+ "loss": 4.2504,
3398
+ "step": 10520
3399
+ },
3400
+ {
3401
+ "epoch": 5.25,
3402
+ "learning_rate": 1.2360393398899816e-05,
3403
+ "loss": 4.2813,
3404
+ "step": 10540
3405
+ },
3406
+ {
3407
+ "epoch": 5.26,
3408
+ "learning_rate": 1.2193698949824972e-05,
3409
+ "loss": 4.2775,
3410
+ "step": 10560
3411
+ },
3412
+ {
3413
+ "epoch": 5.27,
3414
+ "learning_rate": 1.2027004500750126e-05,
3415
+ "loss": 4.2291,
3416
+ "step": 10580
3417
+ },
3418
+ {
3419
+ "epoch": 5.28,
3420
+ "learning_rate": 1.186031005167528e-05,
3421
+ "loss": 4.2709,
3422
+ "step": 10600
3423
+ },
3424
+ {
3425
+ "epoch": 5.29,
3426
+ "learning_rate": 1.1693615602600433e-05,
3427
+ "loss": 4.3188,
3428
+ "step": 10620
3429
+ },
3430
+ {
3431
+ "epoch": 5.3,
3432
+ "learning_rate": 1.1526921153525588e-05,
3433
+ "loss": 4.3012,
3434
+ "step": 10640
3435
+ },
3436
+ {
3437
+ "epoch": 5.31,
3438
+ "learning_rate": 1.1360226704450742e-05,
3439
+ "loss": 4.2261,
3440
+ "step": 10660
3441
+ },
3442
+ {
3443
+ "epoch": 5.32,
3444
+ "learning_rate": 1.1193532255375897e-05,
3445
+ "loss": 4.2179,
3446
+ "step": 10680
3447
+ },
3448
+ {
3449
+ "epoch": 5.33,
3450
+ "learning_rate": 1.1026837806301051e-05,
3451
+ "loss": 4.2907,
3452
+ "step": 10700
3453
+ },
3454
+ {
3455
+ "epoch": 5.34,
3456
+ "learning_rate": 1.0860143357226205e-05,
3457
+ "loss": 4.2918,
3458
+ "step": 10720
3459
+ },
3460
+ {
3461
+ "epoch": 5.35,
3462
+ "learning_rate": 1.0693448908151358e-05,
3463
+ "loss": 4.2488,
3464
+ "step": 10740
3465
+ },
3466
+ {
3467
+ "epoch": 5.36,
3468
+ "learning_rate": 1.0526754459076512e-05,
3469
+ "loss": 4.2927,
3470
+ "step": 10760
3471
+ },
3472
+ {
3473
+ "epoch": 5.37,
3474
+ "learning_rate": 1.0360060010001667e-05,
3475
+ "loss": 4.2602,
3476
+ "step": 10780
3477
+ },
3478
+ {
3479
+ "epoch": 5.38,
3480
+ "learning_rate": 1.0193365560926823e-05,
3481
+ "loss": 4.2836,
3482
+ "step": 10800
3483
+ },
3484
+ {
3485
+ "epoch": 5.38,
3486
+ "eval_validation_accuracy": 1.0,
3487
+ "eval_validation_loss": 5.06640625,
3488
+ "eval_validation_runtime": 337.786,
3489
+ "eval_validation_samples_per_second": 30.579,
3490
+ "eval_validation_steps_per_second": 1.276,
3491
+ "step": 10800
3492
+ },
3493
+ {
3494
+ "epoch": 5.39,
3495
+ "learning_rate": 1.0026671111851977e-05,
3496
+ "loss": 4.2308,
3497
+ "step": 10820
3498
+ },
3499
+ {
3500
+ "epoch": 5.4,
3501
+ "learning_rate": 9.85997666277713e-06,
3502
+ "loss": 4.248,
3503
+ "step": 10840
3504
+ },
3505
+ {
3506
+ "epoch": 5.41,
3507
+ "learning_rate": 9.693282213702284e-06,
3508
+ "loss": 4.2814,
3509
+ "step": 10860
3510
+ },
3511
+ {
3512
+ "epoch": 5.42,
3513
+ "learning_rate": 9.526587764627438e-06,
3514
+ "loss": 4.3033,
3515
+ "step": 10880
3516
+ },
3517
+ {
3518
+ "epoch": 5.43,
3519
+ "learning_rate": 9.359893315552593e-06,
3520
+ "loss": 4.2142,
3521
+ "step": 10900
3522
+ },
3523
+ {
3524
+ "epoch": 5.44,
3525
+ "learning_rate": 9.193198866477747e-06,
3526
+ "loss": 4.2635,
3527
+ "step": 10920
3528
+ },
3529
+ {
3530
+ "epoch": 5.45,
3531
+ "learning_rate": 9.0265044174029e-06,
3532
+ "loss": 4.2951,
3533
+ "step": 10940
3534
+ },
3535
+ {
3536
+ "epoch": 5.46,
3537
+ "learning_rate": 8.859809968328054e-06,
3538
+ "loss": 4.2588,
3539
+ "step": 10960
3540
+ },
3541
+ {
3542
+ "epoch": 5.47,
3543
+ "learning_rate": 8.693115519253209e-06,
3544
+ "loss": 4.3012,
3545
+ "step": 10980
3546
+ },
3547
+ {
3548
+ "epoch": 5.48,
3549
+ "learning_rate": 8.526421070178363e-06,
3550
+ "loss": 4.2285,
3551
+ "step": 11000
3552
+ },
3553
+ {
3554
+ "epoch": 5.49,
3555
+ "learning_rate": 8.359726621103519e-06,
3556
+ "loss": 4.0523,
3557
+ "step": 11020
3558
+ },
3559
+ {
3560
+ "epoch": 5.5,
3561
+ "learning_rate": 8.209701616936157e-06,
3562
+ "loss": 3.2342,
3563
+ "step": 11040
3564
+ },
3565
+ {
3566
+ "epoch": 5.51,
3567
+ "learning_rate": 8.04300716786131e-06,
3568
+ "loss": 3.2146,
3569
+ "step": 11060
3570
+ },
3571
+ {
3572
+ "epoch": 5.52,
3573
+ "learning_rate": 7.876312718786464e-06,
3574
+ "loss": 3.2094,
3575
+ "step": 11080
3576
+ },
3577
+ {
3578
+ "epoch": 5.53,
3579
+ "learning_rate": 7.709618269711619e-06,
3580
+ "loss": 3.203,
3581
+ "step": 11100
3582
+ },
3583
+ {
3584
+ "epoch": 5.54,
3585
+ "learning_rate": 7.542923820636773e-06,
3586
+ "loss": 3.2209,
3587
+ "step": 11120
3588
+ },
3589
+ {
3590
+ "epoch": 5.55,
3591
+ "learning_rate": 7.3762293715619264e-06,
3592
+ "loss": 3.2051,
3593
+ "step": 11140
3594
+ },
3595
+ {
3596
+ "epoch": 5.56,
3597
+ "learning_rate": 7.209534922487081e-06,
3598
+ "loss": 3.2109,
3599
+ "step": 11160
3600
+ },
3601
+ {
3602
+ "epoch": 5.57,
3603
+ "learning_rate": 7.042840473412236e-06,
3604
+ "loss": 3.1812,
3605
+ "step": 11180
3606
+ },
3607
+ {
3608
+ "epoch": 5.58,
3609
+ "learning_rate": 6.87614602433739e-06,
3610
+ "loss": 3.2428,
3611
+ "step": 11200
3612
+ },
3613
+ {
3614
+ "epoch": 5.58,
3615
+ "eval_validation_accuracy": 1.0,
3616
+ "eval_validation_loss": 5.05078125,
3617
+ "eval_validation_runtime": 334.162,
3618
+ "eval_validation_samples_per_second": 30.91,
3619
+ "eval_validation_steps_per_second": 1.29,
3620
+ "step": 11200
3621
+ },
3622
+ {
3623
+ "epoch": 5.59,
3624
+ "learning_rate": 6.709451575262545e-06,
3625
+ "loss": 3.1735,
3626
+ "step": 11220
3627
+ },
3628
+ {
3629
+ "epoch": 5.6,
3630
+ "learning_rate": 6.542757126187698e-06,
3631
+ "loss": 3.1832,
3632
+ "step": 11240
3633
+ },
3634
+ {
3635
+ "epoch": 5.61,
3636
+ "learning_rate": 6.376062677112853e-06,
3637
+ "loss": 3.1974,
3638
+ "step": 11260
3639
+ },
3640
+ {
3641
+ "epoch": 5.62,
3642
+ "learning_rate": 6.209368228038007e-06,
3643
+ "loss": 3.1519,
3644
+ "step": 11280
3645
+ },
3646
+ {
3647
+ "epoch": 5.63,
3648
+ "learning_rate": 6.0426737789631605e-06,
3649
+ "loss": 3.1716,
3650
+ "step": 11300
3651
+ },
3652
+ {
3653
+ "epoch": 5.64,
3654
+ "learning_rate": 5.875979329888315e-06,
3655
+ "loss": 3.1634,
3656
+ "step": 11320
3657
+ },
3658
+ {
3659
+ "epoch": 5.65,
3660
+ "learning_rate": 5.709284880813469e-06,
3661
+ "loss": 3.2217,
3662
+ "step": 11340
3663
+ },
3664
+ {
3665
+ "epoch": 5.66,
3666
+ "learning_rate": 5.5425904317386235e-06,
3667
+ "loss": 3.1975,
3668
+ "step": 11360
3669
+ },
3670
+ {
3671
+ "epoch": 5.67,
3672
+ "learning_rate": 5.375895982663778e-06,
3673
+ "loss": 3.2056,
3674
+ "step": 11380
3675
+ },
3676
+ {
3677
+ "epoch": 5.68,
3678
+ "learning_rate": 5.209201533588931e-06,
3679
+ "loss": 3.1516,
3680
+ "step": 11400
3681
+ },
3682
+ {
3683
+ "epoch": 5.69,
3684
+ "learning_rate": 5.042507084514086e-06,
3685
+ "loss": 3.2109,
3686
+ "step": 11420
3687
+ },
3688
+ {
3689
+ "epoch": 5.7,
3690
+ "learning_rate": 4.87581263543924e-06,
3691
+ "loss": 3.2169,
3692
+ "step": 11440
3693
+ },
3694
+ {
3695
+ "epoch": 5.71,
3696
+ "learning_rate": 4.7091181863643945e-06,
3697
+ "loss": 3.1828,
3698
+ "step": 11460
3699
+ },
3700
+ {
3701
+ "epoch": 5.72,
3702
+ "learning_rate": 4.542423737289549e-06,
3703
+ "loss": 3.1879,
3704
+ "step": 11480
3705
+ },
3706
+ {
3707
+ "epoch": 5.73,
3708
+ "learning_rate": 4.375729288214702e-06,
3709
+ "loss": 3.1868,
3710
+ "step": 11500
3711
+ },
3712
+ {
3713
+ "epoch": 5.74,
3714
+ "learning_rate": 4.209034839139857e-06,
3715
+ "loss": 3.2158,
3716
+ "step": 11520
3717
+ },
3718
+ {
3719
+ "epoch": 5.75,
3720
+ "learning_rate": 4.042340390065011e-06,
3721
+ "loss": 3.1862,
3722
+ "step": 11540
3723
+ },
3724
+ {
3725
+ "epoch": 5.76,
3726
+ "learning_rate": 3.8756459409901646e-06,
3727
+ "loss": 3.1505,
3728
+ "step": 11560
3729
+ },
3730
+ {
3731
+ "epoch": 5.77,
3732
+ "learning_rate": 3.7089514919153198e-06,
3733
+ "loss": 3.1722,
3734
+ "step": 11580
3735
+ },
3736
+ {
3737
+ "epoch": 5.78,
3738
+ "learning_rate": 3.5422570428404737e-06,
3739
+ "loss": 3.1958,
3740
+ "step": 11600
3741
+ },
3742
+ {
3743
+ "epoch": 5.78,
3744
+ "eval_validation_accuracy": 1.0,
3745
+ "eval_validation_loss": 5.046875,
3746
+ "eval_validation_runtime": 332.2974,
3747
+ "eval_validation_samples_per_second": 31.084,
3748
+ "eval_validation_steps_per_second": 1.297,
3749
+ "step": 11600
3750
+ },
3751
+ {
3752
+ "epoch": 5.79,
3753
+ "learning_rate": 3.3755625937656276e-06,
3754
+ "loss": 3.1659,
3755
+ "step": 11620
3756
+ },
3757
+ {
3758
+ "epoch": 5.8,
3759
+ "learning_rate": 3.208868144690782e-06,
3760
+ "loss": 3.1965,
3761
+ "step": 11640
3762
+ },
3763
+ {
3764
+ "epoch": 5.81,
3765
+ "learning_rate": 3.0421736956159364e-06,
3766
+ "loss": 3.2112,
3767
+ "step": 11660
3768
+ },
3769
+ {
3770
+ "epoch": 5.82,
3771
+ "learning_rate": 2.8754792465410903e-06,
3772
+ "loss": 3.1755,
3773
+ "step": 11680
3774
+ },
3775
+ {
3776
+ "epoch": 5.83,
3777
+ "learning_rate": 2.7087847974662442e-06,
3778
+ "loss": 3.201,
3779
+ "step": 11700
3780
+ },
3781
+ {
3782
+ "epoch": 5.84,
3783
+ "learning_rate": 2.542090348391399e-06,
3784
+ "loss": 3.1565,
3785
+ "step": 11720
3786
+ },
3787
+ {
3788
+ "epoch": 5.85,
3789
+ "learning_rate": 2.375395899316553e-06,
3790
+ "loss": 3.1998,
3791
+ "step": 11740
3792
+ },
3793
+ {
3794
+ "epoch": 5.86,
3795
+ "learning_rate": 2.208701450241707e-06,
3796
+ "loss": 3.1801,
3797
+ "step": 11760
3798
+ },
3799
+ {
3800
+ "epoch": 5.87,
3801
+ "learning_rate": 2.0420070011668612e-06,
3802
+ "loss": 3.1721,
3803
+ "step": 11780
3804
+ },
3805
+ {
3806
+ "epoch": 5.88,
3807
+ "learning_rate": 1.8753125520920154e-06,
3808
+ "loss": 3.2096,
3809
+ "step": 11800
3810
+ },
3811
+ {
3812
+ "epoch": 5.89,
3813
+ "learning_rate": 1.7086181030171695e-06,
3814
+ "loss": 3.1563,
3815
+ "step": 11820
3816
+ },
3817
+ {
3818
+ "epoch": 5.9,
3819
+ "learning_rate": 1.5419236539423237e-06,
3820
+ "loss": 3.1879,
3821
+ "step": 11840
3822
+ },
3823
+ {
3824
+ "epoch": 5.91,
3825
+ "learning_rate": 1.375229204867478e-06,
3826
+ "loss": 3.1544,
3827
+ "step": 11860
3828
+ },
3829
+ {
3830
+ "epoch": 5.92,
3831
+ "learning_rate": 1.2085347557926322e-06,
3832
+ "loss": 3.1936,
3833
+ "step": 11880
3834
+ },
3835
+ {
3836
+ "epoch": 5.93,
3837
+ "learning_rate": 1.0418403067177863e-06,
3838
+ "loss": 3.193,
3839
+ "step": 11900
3840
+ },
3841
+ {
3842
+ "epoch": 5.94,
3843
+ "learning_rate": 8.751458576429405e-07,
3844
+ "loss": 3.1856,
3845
+ "step": 11920
3846
+ },
3847
+ {
3848
+ "epoch": 5.95,
3849
+ "learning_rate": 7.084514085680947e-07,
3850
+ "loss": 3.1597,
3851
+ "step": 11940
3852
+ },
3853
+ {
3854
+ "epoch": 5.96,
3855
+ "learning_rate": 5.417569594932489e-07,
3856
+ "loss": 3.1776,
3857
+ "step": 11960
3858
+ },
3859
+ {
3860
+ "epoch": 5.97,
3861
+ "learning_rate": 3.750625104184031e-07,
3862
+ "loss": 3.1353,
3863
+ "step": 11980
3864
+ },
3865
+ {
3866
+ "epoch": 5.98,
3867
+ "learning_rate": 2.0836806134355724e-07,
3868
+ "loss": 3.1876,
3869
+ "step": 12000
3870
+ },
3871
+ {
3872
+ "epoch": 5.98,
3873
+ "eval_validation_accuracy": 1.0,
3874
+ "eval_validation_loss": 5.04296875,
3875
+ "eval_validation_runtime": 336.5674,
3876
+ "eval_validation_samples_per_second": 30.689,
3877
+ "eval_validation_steps_per_second": 1.281,
3878
+ "step": 12000
3879
+ },
3880
+ {
3881
+ "epoch": 5.98,
3882
+ "step": 12000,
3883
+ "total_flos": 1.2861454072393236e+19,
3884
+ "train_loss": 0.2671610107421875,
3885
+ "train_runtime": 4071.6945,
3886
+ "train_samples_per_second": 282.929,
3887
+ "train_steps_per_second": 2.947
3888
+ }
3889
+ ],
3890
+ "max_steps": 12000,
3891
+ "num_train_epochs": 6,
3892
+ "total_flos": 1.2861454072393236e+19,
3893
+ "trial_name": null,
3894
+ "trial_params": null
3895
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44bed58a0e3386ad16e3c11eee3ae6a5b71f6a6f6871f487ee2739c9d839b4bb
3
+ size 4975