| { | |
| "_name_or_path": "/large_storage/goodarzilab/userspace/mohsen/biofm/saved_models/DeCodon/CodonGPT_L2048_l12_a16_b32_r5e-05_wd0.01_g2_euk_adapt_taxid-yzhnq4du/checkpoint-60000", | |
| "architectures": [ | |
| "DeCodon" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "attention_type": "self", | |
| "auto_map": { | |
| "AutoConfig": "configuration_decodon.DeCodonConfig", | |
| "AutoModelForCausalLM": "modeling_decodon.DeCodon", | |
| "AutoTokenizer": "cdsfm.decodon.tokenization_decodon.DeCodonTokenizer" | |
| }, | |
| "classifier_dropout": 0.1, | |
| "dilation_rates": null, | |
| "gamma_init": 1.782709687623856, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 1024, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "is_decoder": true, | |
| "layer_norm_eps": 1e-12, | |
| "lm_type": "distilled", | |
| "max_position_embeddings": 2048, | |
| "model_type": "decodon", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 12, | |
| "pad_token_id": 3, | |
| "position_embedding_type": "rotary", | |
| "rotary_theta": 10000.0, | |
| "segment_lengths": null, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.44.2", | |
| "type_vocab_size": 2, | |
| "use_cache": true, | |
| "use_flash_attn": true, | |
| "use_rotary_emb": true, | |
| "vocab_size": 1736 | |
| } | |