| { | |
| "_name_or_path": "MILVLG/Imp-v1.5-4B-Phi3", | |
| "activation_function": "gelu_new", | |
| "architectures": [ | |
| "ImpPhi3ForCausalLM" | |
| ], | |
| "attn_pdrop": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_imp.ImpPhi3Config", | |
| "AutoModelForCausalLM": "modeling_imp.ImpPhi3ForCausalLM" | |
| }, | |
| "embd_pdrop": 0.0, | |
| "bos_token_id": 1, | |
| "eos_token_id": 32007, | |
| "flash_attn": false, | |
| "flash_rotary": false, | |
| "freeze_mm_mlp_adapter": false, | |
| "hidden_act": "silu", | |
| "hidden_size": 3072, | |
| "image_aspect_ratio": "pad", | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8192, | |
| "max_position_embeddings": 4096, | |
| "mm_hidden_size": 1152, | |
| "mm_projector_lr": 2e-05, | |
| "mm_projector_type": "mlp2x_gelu", | |
| "mm_use_im_patch_token": false, | |
| "mm_use_im_start_end": false, | |
| "mm_vision_select_feature": "patch", | |
| "mm_vision_select_layer": -2, | |
| "mm_vision_tower": "google/siglip-so400m-patch14-384", | |
| "model_type": "imp_phi3", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "original_max_position_embeddings": 4096, | |
| "pad_token_id": 32000, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "sliding_window": 2047, | |
| "tie_word_embeddings": false, | |
| "tokenizer_model_max_length": 2560, | |
| "tokenizer_padding_side": "right", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.36.0", | |
| "tune_mm_mlp_adapter": false, | |
| "use_cache": true, | |
| "use_mm_proj": true, | |
| "fused_dense": false, | |
| "image_token": "<image>", | |
| "image_token_index": 32011, | |
| "img_processor": null, | |
| "layer_norm_epsilon": 1e-05, | |
| "n_embd": 2560, | |
| "n_head": 32, | |
| "n_head_kv": null, | |
| "n_inner": null, | |
| "n_layer": 32, | |
| "n_positions": 3072, | |
| "rotary_dim": 32, | |
| "vision_tower_config": { | |
| "attention_dropout": 0.0, | |
| "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_channels": 3, | |
| "num_hidden_layers": 27, | |
| "patch_size": 14 | |
| }, | |
| "vocab_size": 32064 | |
| } | |