65 lines
1.3 KiB
JSON
65 lines
1.3 KiB
JSON
|
{
|
||
|
|
"architectures": [
|
||
|
|
"LlavaNextForConditionalGeneration"
|
||
|
|
],
|
||
|
|
"ignore_index": -100,
|
||
|
|
"image_grid_pinpoints": [
|
||
|
|
[
|
||
|
|
336,
|
||
|
|
672
|
||
|
|
],
|
||
|
|
[
|
||
|
|
672,
|
||
|
|
336
|
||
|
|
],
|
||
|
|
[
|
||
|
|
672,
|
||
|
|
672
|
||
|
|
],
|
||
|
|
[
|
||
|
|
1008,
|
||
|
|
336
|
||
|
|
],
|
||
|
|
[
|
||
|
|
336,
|
||
|
|
1008
|
||
|
|
]
|
||
|
|
],
|
||
|
|
"image_token_index": 128256,
|
||
|
|
"model_type": "llava_next",
|
||
|
|
"projector_hidden_act": "gelu",
|
||
|
|
"text_config": {
|
||
|
|
"_name_or_path": "models/Meta-Llama-3-8B",
|
||
|
|
"architectures": [
|
||
|
|
"LlamaForCausalLM"
|
||
|
|
],
|
||
|
|
"bos_token_id": 128000,
|
||
|
|
"eos_token_id": 128009,
|
||
|
|
"intermediate_size": 14336,
|
||
|
|
"max_position_embeddings": 8192,
|
||
|
|
"model_type": "llama",
|
||
|
|
"num_key_value_heads": 8,
|
||
|
|
"rms_norm_eps": 1e-05,
|
||
|
|
"rope_theta": 500000.0,
|
||
|
|
"torch_dtype": "bfloat16",
|
||
|
|
"vocab_size": 128320
|
||
|
|
},
|
||
|
|
"tie_word_embeddings": false,
|
||
|
|
"torch_dtype": "float16",
|
||
|
|
"transformers_version": "4.44.0",
|
||
|
|
"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"
|
||
|
|
}
|