| { | |
| "architectures": ["MistralForCausalLM"], | |
| "model_type": "mistral", | |
| "hidden_size": 4096, | |
| "intermediate_size": 11008, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "vocab_size": 50272, | |
| "max_position_embeddings": 2048, | |
| "rotary_embedding_base": 10000, | |
| "tie_word_embeddings": false, | |
| "hidden_act": "silu", | |
| "initializer_range": 0.02, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "pad_token_id": 0, | |
| "torch_dtype": "float16", | |
| "use_cache": true, | |
| "quantization_config": { | |
| "bits": 4, | |
| "group_size": 128, | |
| "dtype": "q4_k_m" | |
| }, | |
| "model_revision": "v0.1", | |
| "instructions": "This model is fine-tuned for instruction-following tasks using the Mistral 7B model.", | |
| "repository": "Aditya757864/Mistral7B", | |
| "version": "0.1" | |
| } | |