samiyousef/Command-R-4-bit
026
1{2 "_name_or_path": "CohereForAI/c4ai-command-r-v01",3 "architectures": [4 "CohereForCausalLM"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "bos_token_id": 5,9 "eos_token_id": 255001,10 "hidden_act": "silu",11 "hidden_size": 8192,12 "initializer_range": 0.02,13 "intermediate_size": 22528,14 "layer_norm_eps": 1e-05,15 "logit_scale": 0.0625,16 "max_position_embeddings": 8192,17 "model_type": "cohere",18 "num_attention_heads": 64,19 "num_hidden_layers": 40,20 "num_key_value_heads": 64,21 "pad_token_id": 0,22 "pretraining_tp": 1,23 "quantization_config": {24 "_load_in_4bit": true,25 "_load_in_8bit": false,26 "bnb_4bit_compute_dtype": "float16",27 "bnb_4bit_quant_storage": "uint8",28 "bnb_4bit_quant_type": "fp4",29 "bnb_4bit_use_double_quant": false,30 "llm_int8_enable_fp32_cpu_offload": false,31 "llm_int8_has_fp16_weight": false,32 "llm_int8_skip_modules": null,33 "llm_int8_threshold": 6.0,34 "load_in_4bit": true,35 "load_in_8bit": false,36 "quant_method": "bitsandbytes"37 },38 "rope_theta": 8000000.0,39 "torch_dtype": "float16",40 "transformers_version": "4.39.0.dev0",41 "use_cache": true,42 "vocab_size": 25600043}44 