101 lines
2.1 KiB
JSON
101 lines
2.1 KiB
JSON
{
|
|
"_name_or_path": "/home/goramesh/local/gramesh/Instella-VL-1B/",
|
|
"architectures": [
|
|
"InstellaVLForCausalLM"
|
|
],
|
|
"auto_map": {
|
|
"AutoConfig": "modeling_instellavl.InstellaVLConfig",
|
|
"AutoModelForCausalLM": "modeling_instellavl.InstellaVLForCausalLM"
|
|
},
|
|
"attention_bias": false,
|
|
"attention_dropout": 0.0,
|
|
"clip_qkv": null,
|
|
"eos_token_id": 50279,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 2048,
|
|
"image_aspect_ratio": "anyres",
|
|
"image_crop_resolution": null,
|
|
"image_grid_pinpoints": [
|
|
[
|
|
336,
|
|
336
|
|
],
|
|
[
|
|
336,
|
|
672
|
|
],
|
|
[
|
|
336,
|
|
1008
|
|
],
|
|
[
|
|
336,
|
|
1344
|
|
],
|
|
[
|
|
336,
|
|
1680
|
|
],
|
|
[
|
|
672,
|
|
336
|
|
],
|
|
[
|
|
672,
|
|
672
|
|
],
|
|
[
|
|
1008,
|
|
336
|
|
],
|
|
[
|
|
1344,
|
|
336
|
|
],
|
|
[
|
|
1680,
|
|
336
|
|
]
|
|
],
|
|
"image_split_resolution": null,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 8192,
|
|
"max_position_embeddings": 2048,
|
|
"mm_anyres_choose_method": "best_fit",
|
|
"mm_compact_visual_tokens": false,
|
|
"mm_downsample_ratio": 1,
|
|
"mm_hidden_size": 1024,
|
|
"mm_newline_position": "one_token",
|
|
"mm_patch_merge_type": "spatial_unpad",
|
|
"mm_projector_lr": null,
|
|
"mm_projector_type": "mlp2x_gelu",
|
|
"mm_resampler_type": null,
|
|
"mm_spatial_pool_mode": "bilinear",
|
|
"mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
|
|
"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",
|
|
"mm_vision_tower_lr": null,
|
|
"model_type": "instellavl",
|
|
"num_attention_heads": 16,
|
|
"num_hidden_layers": 16,
|
|
"num_key_value_heads": 16,
|
|
"online_training": true,
|
|
"pad_token_id": 1,
|
|
"pos_skipping_range": 4096,
|
|
"rope_scaling": null,
|
|
"rope_theta": 10000.0,
|
|
"tie_word_embeddings": true,
|
|
"tokenizer_model_max_length": 32768,
|
|
"tokenizer_padding_side": "right",
|
|
"torch_dtype": "float16",
|
|
"transformers_version": "4.45.1",
|
|
"use_cache": true,
|
|
"use_mm_proj": true,
|
|
"use_pos_skipping": false,
|
|
"vision_tower_pretrained": null,
|
|
"vocab_size": 50282
|
|
}
|