CloneBO/CloneLM-Heavy
15.7k
1{2 "_name_or_path": "/scratch/nvg7279/bighat_temp_heavy",3 "architectures": [4 "MistralForCausalLM"5 ],6 "attention_dropout": 0.0,7 "bos_token_id": 1,8 "eos_token_id": 2,9 "head_dim": 64,10 "hidden_act": "silu",11 "hidden_size": 1024,12 "initializer_range": 0.02,13 "intermediate_size": 4096,14 "max_position_embeddings": 2048,15 "model_type": "mistral",16 "num_attention_heads": 16,17 "num_hidden_layers": 24,18 "num_key_value_heads": 8,19 "rms_norm_eps": 1e-05,20 "rope_theta": 10000.0,21 "sliding_window": 4096,22 "tie_word_embeddings": false,23 "torch_dtype": "float32",24 "transformers_version": "4.46.3",25 "use_cache": true,26 "vocab_size": 2927}28 