Llava-v1.5-7B-hf / config.json
shauray's picture
Update config.json
e8c5c55
{
"_commit_hash": null,
"_name_or_path": "shauray/Llava-v1.5-7B-hf",
"model_type": "llava",
"architectures": [
"LlavaForCausalLM"
],
"text_config": {
"_name_or_path": "",
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 11008,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 32,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.32.0.dev0",
"use_cache": true,
"vocab_size": 32000
},
"vision_config": {
"_name_or_path": "",
"bos_token_id": 1,
"eos_token_id": 2,
"freeze_mm_mlp_adapter": true,
"hidden_act": "silu",
"proj_hidden_size": 4096,
"image_aspect_ratio": "square",
"image_grid_pinpoints": null,
"initializer_range": 0.02,
"intermediate_size": 11008,
"max_position_embeddings": 4096,
"mm_hidden_size": 1024,
"mm_resampler_type": null,
"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_vision",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 32,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.31.0",
"tune_mm_mlp_adapter": false,
"tune_mm_vision_resampler": false,
"use_cache": false,
"use_mm_proj": true,
"vocab_size": 32000,
"projector": "Linear",
"hidden_size":768,
"intermediate_size":4096,
"projection_dim":768,
"num_hidden_layers":12,
"num_attention_heads":12,
"num_channels":3,
"image_size":336,
"patch_size":14,
"hidden_act":"quick_gelu",
"layer_norm_eps":1e-5,
"attention_dropout":0.0,
"initializer_range":0.02,
"initializer_factor":1.0
}
}