llava-next-docviz / config.json
ShinDJ's picture
Upload LlavaNextForConditionalGeneration
b7723db verified
raw
history blame
1.61 kB
{
"_name_or_path": "/home/work/hdd_data/VLM/llava-next/results/doc-stage2/checkpoint-20904",
"architectures": [
"LlavaNextForConditionalGeneration"
],
"ignore_index": -100,
"image_grid_pinpoints": [
[
336,
672
],
[
672,
672
]
],
"image_seq_length": 576,
"image_token_index": 128256,
"model_type": "llava_next",
"projector_hidden_act": "gelu",
"text_config": {
"_attn_implementation_autoset": true,
"_name_or_path": "MLP-KTLim/llama-3.1-Korean-Bllossom-8B",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 128000,
"eos_token_id": 128009,
"intermediate_size": 14336,
"max_position_embeddings": 131072,
"model_type": "llama",
"num_key_value_heads": 8,
"rms_norm_eps": 1e-05,
"rope_scaling": {
"factor": 8.0,
"high_freq_factor": 4.0,
"low_freq_factor": 1.0,
"original_max_position_embeddings": 8192,
"rope_type": "llama3"
},
"rope_theta": 500000.0,
"torch_dtype": "bfloat16",
"use_cache": false,
"vocab_size": 128257
},
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.46.2",
"use_image_newline_parameter": true,
"vision_config": {
"hidden_size": 1024,
"image_size": 336,
"intermediate_size": 4096,
"model_type": "clip_vision_model",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"patch_size": 14,
"projection_dim": 768,
"vocab_size": 32000
},
"vision_feature_layer": -2,
"vision_feature_select_strategy": "default",
"vocab_size": 32064
}