| { | |
| "_name_or_path": "mistralai/Mistral-7B-v0.1", | |
| "architectures": [ | |
| "MistralForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "id2label": { | |
| "0": "loc1lay1", | |
| "1": "loc1lay2", | |
| "2": "loc1lay3", | |
| "3": "loc1lay4", | |
| "4": "loc2lay1", | |
| "5": "loc2lay2", | |
| "6": "loc2lay3", | |
| "7": "loc2lay4", | |
| "8": "loc3lay1", | |
| "9": "loc3lay2", | |
| "10": "loc3lay3", | |
| "11": "loc3lay4", | |
| "12": "loc4lay1", | |
| "13": "loc4lay2", | |
| "14": "loc4lay3", | |
| "15": "loc4lay4", | |
| "16": "loc5+", | |
| "17": "loc1", | |
| "18": "loc2", | |
| "19": "loc3", | |
| "20": "loc4", | |
| "21": "nfw", | |
| "22": "tfw", | |
| "23": "tfwc", | |
| "24": "tfwp", | |
| "25": "lay1", | |
| "26": "lay2", | |
| "27": "lay3", | |
| "28": "lay4" | |
| }, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 14336, | |
| "label2id": { | |
| "lay1": 25, | |
| "lay2": 26, | |
| "lay3": 27, | |
| "lay4": 28, | |
| "loc1": 17, | |
| "loc1lay1": 0, | |
| "loc1lay2": 1, | |
| "loc1lay3": 2, | |
| "loc1lay4": 3, | |
| "loc2": 18, | |
| "loc2lay1": 4, | |
| "loc2lay2": 5, | |
| "loc2lay3": 6, | |
| "loc2lay4": 7, | |
| "loc3": 19, | |
| "loc3lay1": 8, | |
| "loc3lay2": 9, | |
| "loc3lay3": 10, | |
| "loc3lay4": 11, | |
| "loc4": 20, | |
| "loc4lay1": 12, | |
| "loc4lay2": 13, | |
| "loc4lay3": 14, | |
| "loc4lay4": 15, | |
| "loc5+": 16, | |
| "nfw": 21, | |
| "tfw": 22, | |
| "tfwc": 23, | |
| "tfwp": 24 | |
| }, | |
| "max_position_embeddings": 32768, | |
| "model_type": "mistral", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 8, | |
| "rms_norm_eps": 1e-05, | |
| "rope_theta": 10000.0, | |
| "sliding_window": 4096, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.40.1", | |
| "use_cache": true, | |
| "vocab_size": 32000 | |
| } | |