Upload config.json with huggingface_hub
Browse files- config.json +88 -0
config.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwenForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 151643,
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 3584,
|
| 14 |
+
"image_aspect_ratio": "pad",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 18944,
|
| 20 |
+
"layer_types": [
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 28,
|
| 52 |
+
"mm_hidden_size": 1152,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_pool_size": 2048,
|
| 56 |
+
"mm_projector_lr": null,
|
| 57 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 58 |
+
"mm_resampler_type": null,
|
| 59 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 60 |
+
"mm_spatial_pool_stride": null,
|
| 61 |
+
"mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
|
| 62 |
+
"mm_use_im_patch_token": false,
|
| 63 |
+
"mm_use_im_start_end": false,
|
| 64 |
+
"mm_vision_select_feature": "patch",
|
| 65 |
+
"mm_vision_select_layer": -2,
|
| 66 |
+
"mm_vision_tower": "SigLIP2/siglip2-so400m-patch14-384",
|
| 67 |
+
"mm_vision_tower_lr": 2e-06,
|
| 68 |
+
"model_type": "qwen2",
|
| 69 |
+
"num_attention_heads": 28,
|
| 70 |
+
"num_hidden_layers": 28,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": false,
|
| 78 |
+
"tokenizer_model_max_length": 321120,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"torch_dtype": "bfloat16",
|
| 81 |
+
"transformers_version": "4.53.0.dev0",
|
| 82 |
+
"use_cache": true,
|
| 83 |
+
"use_mm_proj": true,
|
| 84 |
+
"use_pos_skipping": false,
|
| 85 |
+
"use_sliding_window": false,
|
| 86 |
+
"vision_tower_pretrained": null,
|
| 87 |
+
"vocab_size": 152064
|
| 88 |
+
}
|