nanoLLaVA / config.json
Xenova's picture
Xenova HF staff
Add `num_image_tokens` to config.json (#2)
695e2e9 verified
raw
history blame
2.35 kB
{
"_name_or_path": "/home/azureuser/nanoLLaVA/checkpoint-5426",
"architectures": [
"LlavaQwen2ForCausalLM"
],
"auto_map": {
"AutoConfig": "configuration_llava_qwen2.LlavaQwen2Config",
"AutoModelForCausalLM": "modeling_llava_qwen2.LlavaQwen2ForCausalLM"
},
"attention_dropout": 0.0,
"bos_token_id": 151645,
"eos_token_id": 151645,
"freeze_mm_mlp_adapter": false,
"hidden_act": "silu",
"hidden_size": 1024,
"image_aspect_ratio": "pad",
"initializer_range": 0.02,
"intermediate_size": 2816,
"max_position_embeddings": 32768,
"max_window_layers": 21,
"mm_hidden_size": 1152,
"mm_projector_lr": null,
"mm_projector_type": "mlp2x_gelu",
"mm_vision_tower": "google/siglip-so400m-patch14-384",
"language_model": "vilm/Quyen-SE-v0.1",
"model_type": "llava",
"text_config": {
"_name_or_path": "vilm/Quyen-SE-v0.1",
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 2816,
"max_position_embeddings": 32768,
"max_window_layers": 21,
"model_type": "qwen2",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"num_image_tokens": 729,
"num_key_value_heads": 16,
"rms_norm_eps": 1e-06,
"rope_theta": 1000000.0,
"sliding_window": 4096,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.38.1",
"use_cache": false,
"use_sliding_window": false,
"vocab_size": 151936
},
"vision_config": {
"hidden_size": 1152,
"image_size": 384,
"intermediate_size": 4304,
"model_type": "siglip_vision_model",
"num_attention_heads": 16,
"num_hidden_layers": 27,
"num_image_tokens": 729,
"patch_size": 14
},
"num_attention_heads": 16,
"num_hidden_layers": 24,
"num_image_tokens": 729,
"num_key_value_heads": 16,
"rms_norm_eps": 1e-06,
"rope_theta": 1000000.0,
"sliding_window": 4096,
"tie_word_embeddings": false,
"tokenizer_model_max_length": 4096,
"tokenizer_padding_side": "right",
"torch_dtype": "bfloat16",
"transformers_version": "4.39.2",
"tune_mm_mlp_adapter": false,
"use_cache": false,
"use_mm_proj": true,
"use_sliding_window": false,
"vocab_size": 151936,
"image_token_index": -200
}