LLaVAR-2-Phi3-mini / config.json
smz8599's picture
commit files to HF hub
cbef705
{
"_name_or_path": "microsoft/Phi-3-mini-4k-instruct",
"architectures": [
"LlavaPhiForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "microsoft/Phi-3-mini-4k-instruct--configuration_phi3.Phi3Config",
"AutoModelForCausalLM": "microsoft/Phi-3-mini-4k-instruct--modeling_phi3.Phi3ForCausalLM"
},
"bos_token_id": 1,
"delay_load": false,
"embd_pdrop": 0.0,
"eos_token_id": 32000,
"freeze_mm_mlp_adapter": false,
"freeze_vision": false,
"hidden_act": "silu",
"hidden_size": 3072,
"image_aspect_ratio": "pad",
"image_grid_pinpoints": null,
"initializer_range": 0.02,
"input_image_size": 768,
"intermediate_size": 8192,
"is_multipath_encoder": true,
"max_position_embeddings": 4096,
"mm_hidden_size": 1024,
"mm_patch_merge_type": "flat",
"mm_projector_lr": null,
"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": "openai/clip-vit-large-patch14-336",
"model_type": "llava_phi",
"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": 4096,
"tokenizer_padding_side": "right",
"torch_dtype": "bfloat16",
"transformers_version": "4.42.0.dev0",
"tune_mm_mlp_adapter": false,
"use_cache": true,
"use_mm_proj": true,
"vision_tower_slow": "convnext_large_mlp.clip_laion2b_ft_320",
"vocab_size": 32064
}