{ "_name_or_path": "microsoft/phi-1_5", "architectures": [ "PhiForCausalLM" ], "attention_dropout": 0.0, "attn_level": "patch_layer", "bos_token_id": null, "cap_softmax": false, "cap_softmax_value": 0.06, "confidence_gate": false, "confidence_gate_slope": 10.0, "confidence_gate_threshold": 0.6, "embd_pdrop": 0.0, "eos_token_id": null, "force_reserve_layer": true, "freeze_mm_mlp_adapter": false, "freeze_residual_scaling_coef": false, "hidden_act": "gelu_new", "hidden_size": 2048, "hybrid_beta_init": null, "hybrid_beta_override": null, "hybrid_disagreement_metric": "kl_local_global", "hybrid_reserve_mode": "local_gate", "image_aspect_ratio": "pad", "initializer_range": 0.02, "intermediate_size": 8192, "layer_norm_eps": 1e-05, "learnable_gate": false, "max_position_embeddings": 2048, "mm_hidden_size": 1152, "mm_patch_merge_type": "flat", "mm_projector_lr": 5e-05, "mm_projector_type": "mlp2x_gelu", "mm_text_hidden_size": 1152, "mm_use_im_patch_token": false, "mm_use_im_start_end": false, "mm_use_text_enc": true, "mm_vision_select_feature": "patch", "mm_vision_select_layer": -2, "mm_vision_tower": "google/siglip-so400m-patch14-384", "model_type": "phi", "num_attention_heads": 32, "num_hidden_layers": 24, "num_key_value_heads": 32, "num_layers_to_choose": 1, "pad_token_id": 50256, "partial_rotary_factor": 0.5, "qk_layernorm": false, "reserve_layer": -2, "resid_pdrop": 0.0, "rope_scaling": null, "rope_theta": 10000.0, "router_lr": 5e-05, "save_router_stats": true, "save_steps_router_stats": 5, "text_attention_dim": 1024, "text_attention_type": "qk", "text_conditioning": "attention", "text_pool": "mean", "text_scale": 1.0, "tie_word_embeddings": false, "tokenizer_model_max_length": 2048, "tokenizer_padding_side": "right", "torch_dtype": "float16", "transformers_version": "4.37.2", "tune_mm_mlp_adapter": false, "tune_residual_scaling_coef": false, "use_cache": true, "use_mm_proj": true, "use_text_conditioned_router": true, "use_text_only_gating": false, "vision_tower_delta": false, "vocab_size": 51200 }