| { | |
| "_name_or_path": "/data5/models//llama-joycaption-alpha-two-hf-llava", | |
| "architectures": [ | |
| "LlavaForConditionalGeneration" | |
| ], | |
| "ignore_index": -100, | |
| "image_seq_length": 729, | |
| "image_token_index": 128077, | |
| "model_type": "llava", | |
| "projector_hidden_act": "gelu", | |
| "quantization_config": { | |
| "amp": true, | |
| "autoround_version": "0.4.2.dev", | |
| "backend": "auto_round:gptq:exllamav2", | |
| "batch_size": 8, | |
| "bits": 4, | |
| "data_type": "int", | |
| "dataset": "NeelNanda/pile-10k", | |
| "enable_minmax_tuning": true, | |
| "enable_norm_bias_tuning": false, | |
| "enable_quanted_input": true, | |
| "gradient_accumulate_steps": 1, | |
| "group_size": 128, | |
| "iters": 1000, | |
| "low_gpu_mem_usage": false, | |
| "lr": 0.001, | |
| "minmax_lr": 0.001, | |
| "nsamples": 512, | |
| "quant_method": "intel/auto-round", | |
| "scale_dtype": "torch.float16", | |
| "seqlen": 2048, | |
| "sym": true, | |
| "to_quant_block_names": "language_model.model.layers" | |
| }, | |
| "text_config": { | |
| "_name_or_path": "unsloth/Meta-Llama-3.1-8B-Instruct", | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "bos_token_id": 128000, | |
| "eos_token_id": [ | |
| 128001, | |
| 128008, | |
| 128009 | |
| ], | |
| "intermediate_size": 14336, | |
| "max_position_embeddings": 131072, | |
| "model_type": "llama", | |
| "num_key_value_heads": 8, | |
| "pad_token_id": 128004, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": { | |
| "factor": 8.0, | |
| "high_freq_factor": 4.0, | |
| "low_freq_factor": 1.0, | |
| "original_max_position_embeddings": 8192, | |
| "rope_type": "llama3" | |
| }, | |
| "rope_theta": 500000.0, | |
| "torch_dtype": "bfloat16", | |
| "unsloth_version": "2024.9", | |
| "vocab_size": 128256 | |
| }, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.44.1", | |
| "vision_config": { | |
| "_name_or_path": "google/siglip-so400m-patch14-384", | |
| "architectures": [ | |
| "SiglipVisionModel" | |
| ], | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 1152, | |
| "image_size": 384, | |
| "intermediate_size": 4304, | |
| "layer_norm_eps": 1e-06, | |
| "model_type": "siglip_vision_model", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 27, | |
| "patch_size": 14, | |
| "torch_dtype": "float32" | |
| }, | |
| "vision_feature_layer": -2, | |
| "vision_feature_select_strategy": "full" | |
| } | |