| { | |
| "_name_or_path": "/mnt/beegfs/mixed-tier/share/projects/fastdraft/project_finetuned_models/phi3-base-50M_fineweb5BT_CP_5BTcode_2.5BTtext/combi_alpaca_oig_magpie_code-evol__alpaca", | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 1, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 32000, | |
| "hidden_act": "silu", | |
| "hidden_size": 512, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 1408, | |
| "max_position_embeddings": 4096, | |
| "model_type": "phi3", | |
| "num_attention_heads": 8, | |
| "num_hidden_layers": 6, | |
| "num_key_value_heads": 8, | |
| "original_max_position_embeddings": 4096, | |
| "pad_token_id": 32000, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "sliding_window": 2047, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "4.44.0", | |
| "use_cache": false, | |
| "vocab_size": 32064 | |
| } |