Upload model trained with Unsloth
Browse filesUpload model trained with Unsloth 2x faster
- config.json +197 -185
- generation_config.json +7 -0
- model.safetensors +3 -0
config.json
CHANGED
|
@@ -1,212 +1,224 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
-
|
| 4 |
],
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"dtype": "float16",
|
| 7 |
"eos_token_id": 151645,
|
| 8 |
"hidden_act": "silu",
|
| 9 |
"hidden_size": 1536,
|
| 10 |
-
"image_token_id":
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"intermediate_size": 8960,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"max_position_embeddings": 32768,
|
| 14 |
"max_window_layers": 28,
|
| 15 |
-
"model_type": "
|
| 16 |
"num_attention_heads": 12,
|
| 17 |
"num_hidden_layers": 28,
|
| 18 |
"num_key_value_heads": 2,
|
| 19 |
"pad_token_id": 151654,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
"rms_norm_eps": 1e-06,
|
| 21 |
"rope_scaling": {
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
},
|
| 30 |
"rope_theta": 1000000.0,
|
| 31 |
-
"sliding_window":
|
| 32 |
-
"
|
| 33 |
-
"_name_or_path": "unsloth/Qwen2-VL-2B-Instruct",
|
| 34 |
-
"architectures": [
|
| 35 |
-
"Qwen2VLForConditionalGeneration"
|
| 36 |
-
],
|
| 37 |
-
"attention_dropout": 0.0,
|
| 38 |
-
"dtype": "float16",
|
| 39 |
-
"eos_token_id": 151645,
|
| 40 |
-
"hidden_act": "silu",
|
| 41 |
-
"hidden_size": 1536,
|
| 42 |
-
"image_token_id": null,
|
| 43 |
-
"initializer_range": 0.02,
|
| 44 |
-
"intermediate_size": 8960,
|
| 45 |
-
"layer_types": [
|
| 46 |
-
"full_attention",
|
| 47 |
-
"full_attention",
|
| 48 |
-
"full_attention",
|
| 49 |
-
"full_attention",
|
| 50 |
-
"full_attention",
|
| 51 |
-
"full_attention",
|
| 52 |
-
"full_attention",
|
| 53 |
-
"full_attention",
|
| 54 |
-
"full_attention",
|
| 55 |
-
"full_attention",
|
| 56 |
-
"full_attention",
|
| 57 |
-
"full_attention",
|
| 58 |
-
"full_attention",
|
| 59 |
-
"full_attention",
|
| 60 |
-
"full_attention",
|
| 61 |
-
"full_attention",
|
| 62 |
-
"full_attention",
|
| 63 |
-
"full_attention",
|
| 64 |
-
"full_attention",
|
| 65 |
-
"full_attention",
|
| 66 |
-
"full_attention",
|
| 67 |
-
"full_attention",
|
| 68 |
-
"full_attention",
|
| 69 |
-
"full_attention",
|
| 70 |
-
"full_attention",
|
| 71 |
-
"full_attention",
|
| 72 |
-
"full_attention",
|
| 73 |
-
"full_attention"
|
| 74 |
-
],
|
| 75 |
-
"max_position_embeddings": 32768,
|
| 76 |
-
"max_window_layers": 28,
|
| 77 |
-
"model_type": "qwen2_vl_text",
|
| 78 |
-
"num_attention_heads": 12,
|
| 79 |
-
"num_hidden_layers": 28,
|
| 80 |
-
"num_key_value_heads": 2,
|
| 81 |
-
"pad_token_id": 151654,
|
| 82 |
-
"quantization_config": {
|
| 83 |
-
"_load_in_4bit": true,
|
| 84 |
-
"_load_in_8bit": false,
|
| 85 |
-
"bnb_4bit_compute_dtype": "bfloat16",
|
| 86 |
-
"bnb_4bit_quant_storage": "uint8",
|
| 87 |
-
"bnb_4bit_quant_type": "nf4",
|
| 88 |
-
"bnb_4bit_use_double_quant": true,
|
| 89 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
| 90 |
-
"llm_int8_has_fp16_weight": false,
|
| 91 |
-
"llm_int8_skip_modules": [
|
| 92 |
-
"lm_head",
|
| 93 |
-
"multi_modal_projector",
|
| 94 |
-
"merger",
|
| 95 |
-
"modality_projection",
|
| 96 |
-
"model.layers.1.mlp",
|
| 97 |
-
"visual.blocks.29.mlp",
|
| 98 |
-
"visual.blocks.21.mlp",
|
| 99 |
-
"visual.blocks.24.attn",
|
| 100 |
-
"visual.blocks.26.mlp",
|
| 101 |
-
"visual.blocks.23.attn",
|
| 102 |
-
"visual.blocks.18.mlp",
|
| 103 |
-
"visual.blocks.19.mlp",
|
| 104 |
-
"visual.blocks.22.attn",
|
| 105 |
-
"visual.blocks.17.mlp",
|
| 106 |
-
"visual.blocks.25.attn",
|
| 107 |
-
"visual.blocks.24.mlp",
|
| 108 |
-
"visual.blocks.26.attn",
|
| 109 |
-
"visual.blocks.29.attn",
|
| 110 |
-
"visual.blocks.21.attn",
|
| 111 |
-
"visual.blocks.22.mlp",
|
| 112 |
-
"visual.blocks.14.mlp",
|
| 113 |
-
"visual.blocks.23.mlp",
|
| 114 |
-
"visual.blocks.14.attn",
|
| 115 |
-
"visual.merger.mlp",
|
| 116 |
-
"visual.blocks.16.mlp",
|
| 117 |
-
"visual.blocks.20.mlp",
|
| 118 |
-
"visual.blocks.25.mlp",
|
| 119 |
-
"visual.blocks.13.mlp",
|
| 120 |
-
"visual.blocks.18.attn",
|
| 121 |
-
"visual.blocks.20.attn",
|
| 122 |
-
"visual.blocks.17.attn",
|
| 123 |
-
"visual.blocks.15.mlp",
|
| 124 |
-
"visual.blocks.27.attn",
|
| 125 |
-
"visual.blocks.28.attn",
|
| 126 |
-
"visual.blocks.12.mlp",
|
| 127 |
-
"visual.blocks.16.attn",
|
| 128 |
-
"visual.blocks.19.attn",
|
| 129 |
-
"visual.blocks.15.attn",
|
| 130 |
-
"visual.blocks.11.attn",
|
| 131 |
-
"visual.blocks.13.attn",
|
| 132 |
-
"visual.blocks.12.attn",
|
| 133 |
-
"visual.blocks.10.mlp",
|
| 134 |
-
"visual.blocks.11.mlp",
|
| 135 |
-
"visual.blocks.10.attn",
|
| 136 |
-
"visual.blocks.9.mlp",
|
| 137 |
-
"visual.blocks.5.attn",
|
| 138 |
-
"visual.blocks.9.attn",
|
| 139 |
-
"visual.blocks.4.attn",
|
| 140 |
-
"visual.blocks.6.mlp",
|
| 141 |
-
"visual.blocks.7.mlp",
|
| 142 |
-
"visual.blocks.1.attn",
|
| 143 |
-
"visual.blocks.6.attn",
|
| 144 |
-
"visual.blocks.7.attn",
|
| 145 |
-
"visual.blocks.5.mlp",
|
| 146 |
-
"visual.blocks.8.attn",
|
| 147 |
-
"visual.blocks.8.mlp",
|
| 148 |
-
"visual.blocks.4.mlp",
|
| 149 |
-
"visual.blocks.3.attn",
|
| 150 |
-
"visual.blocks.3.mlp",
|
| 151 |
-
"visual.blocks.1.mlp",
|
| 152 |
-
"visual.blocks.2.mlp",
|
| 153 |
-
"visual.blocks.2.attn",
|
| 154 |
-
"visual.blocks.0.attn",
|
| 155 |
-
"visual.blocks.0.mlp",
|
| 156 |
-
"visual.blocks.31.mlp"
|
| 157 |
-
],
|
| 158 |
-
"llm_int8_threshold": 6.0,
|
| 159 |
-
"load_in_4bit": true,
|
| 160 |
-
"load_in_8bit": false,
|
| 161 |
-
"quant_method": "bitsandbytes"
|
| 162 |
-
},
|
| 163 |
-
"rms_norm_eps": 1e-06,
|
| 164 |
-
"rope_scaling": {
|
| 165 |
-
"mrope_section": [
|
| 166 |
-
16,
|
| 167 |
-
24,
|
| 168 |
-
24
|
| 169 |
-
],
|
| 170 |
-
"rope_type": "default",
|
| 171 |
-
"type": "default"
|
| 172 |
-
},
|
| 173 |
-
"rope_theta": 1000000.0,
|
| 174 |
-
"sliding_window": null,
|
| 175 |
-
"tie_word_embeddings": true,
|
| 176 |
-
"unsloth_fixed": true,
|
| 177 |
-
"use_cache": true,
|
| 178 |
-
"use_sliding_window": false,
|
| 179 |
-
"video_token_id": null,
|
| 180 |
-
"vision_end_token_id": 151653,
|
| 181 |
-
"vision_start_token_id": 151652,
|
| 182 |
-
"vision_token_id": 151654,
|
| 183 |
-
"vocab_size": 151936
|
| 184 |
-
},
|
| 185 |
-
"transformers_version": "4.56.2",
|
| 186 |
"unsloth_fixed": true,
|
| 187 |
-
"unsloth_version": "2025.9.11",
|
| 188 |
"use_cache": true,
|
| 189 |
"use_sliding_window": false,
|
| 190 |
-
"video_token_id":
|
| 191 |
-
"vision_config": {
|
| 192 |
-
"depth": 32,
|
| 193 |
-
"dtype": "float16",
|
| 194 |
-
"embed_dim": 1280,
|
| 195 |
-
"hidden_act": "quick_gelu",
|
| 196 |
-
"hidden_size": 1536,
|
| 197 |
-
"in_channels": 3,
|
| 198 |
-
"in_chans": 3,
|
| 199 |
-
"initializer_range": 0.02,
|
| 200 |
-
"mlp_ratio": 4,
|
| 201 |
-
"model_type": "qwen2_vl",
|
| 202 |
-
"num_heads": 16,
|
| 203 |
-
"patch_size": 14,
|
| 204 |
-
"spatial_merge_size": 2,
|
| 205 |
-
"spatial_patch_size": 14,
|
| 206 |
-
"temporal_patch_size": 2
|
| 207 |
-
},
|
| 208 |
"vision_end_token_id": 151653,
|
| 209 |
"vision_start_token_id": 151652,
|
| 210 |
"vision_token_id": 151654,
|
| 211 |
"vocab_size": 151936
|
| 212 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2VLForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"dtype": "float16",
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 1536,
|
| 10 |
+
"image_token_id": 151655,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 28,
|
| 15 |
+
"model_type": "qwen2_vl",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"pad_token_id": 151654,
|
| 20 |
+
"quantization_config": {
|
| 21 |
+
"bnb_4bit_compute_dtype": "float16",
|
| 22 |
+
"bnb_4bit_quant_type": "nf4",
|
| 23 |
+
"bnb_4bit_use_double_quant": true,
|
| 24 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 25 |
+
"llm_int8_has_fp16_weight": false,
|
| 26 |
+
"llm_int8_skip_modules": null,
|
| 27 |
+
"llm_int8_threshold": 6.0,
|
| 28 |
+
"load_in_4bit": true,
|
| 29 |
+
"load_in_8bit": false,
|
| 30 |
+
"quant_method": "bitsandbytes"
|
| 31 |
+
},
|
| 32 |
+
"rms_norm_eps": 1e-06,
|
| 33 |
+
"rope_scaling": {
|
| 34 |
+
"mrope_section": [
|
| 35 |
+
16,
|
| 36 |
+
24,
|
| 37 |
+
24
|
| 38 |
+
],
|
| 39 |
+
"rope_type": "default",
|
| 40 |
+
"type": "default"
|
| 41 |
+
},
|
| 42 |
+
"rope_theta": 1000000.0,
|
| 43 |
+
"sliding_window": 32768,
|
| 44 |
+
"text_config": {
|
| 45 |
+
"_name_or_path": "unsloth/Qwen2-VL-2B-Instruct",
|
| 46 |
"architectures": [
|
| 47 |
+
"Qwen2VLForConditionalGeneration"
|
| 48 |
],
|
| 49 |
"attention_dropout": 0.0,
|
| 50 |
"dtype": "float16",
|
| 51 |
"eos_token_id": 151645,
|
| 52 |
"hidden_act": "silu",
|
| 53 |
"hidden_size": 1536,
|
| 54 |
+
"image_token_id": null,
|
| 55 |
"initializer_range": 0.02,
|
| 56 |
"intermediate_size": 8960,
|
| 57 |
+
"layer_types": [
|
| 58 |
+
"full_attention",
|
| 59 |
+
"full_attention",
|
| 60 |
+
"full_attention",
|
| 61 |
+
"full_attention",
|
| 62 |
+
"full_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"full_attention",
|
| 65 |
+
"full_attention",
|
| 66 |
+
"full_attention",
|
| 67 |
+
"full_attention",
|
| 68 |
+
"full_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"full_attention",
|
| 71 |
+
"full_attention",
|
| 72 |
+
"full_attention",
|
| 73 |
+
"full_attention",
|
| 74 |
+
"full_attention",
|
| 75 |
+
"full_attention",
|
| 76 |
+
"full_attention",
|
| 77 |
+
"full_attention",
|
| 78 |
+
"full_attention",
|
| 79 |
+
"full_attention",
|
| 80 |
+
"full_attention",
|
| 81 |
+
"full_attention",
|
| 82 |
+
"full_attention",
|
| 83 |
+
"full_attention",
|
| 84 |
+
"full_attention",
|
| 85 |
+
"full_attention"
|
| 86 |
+
],
|
| 87 |
"max_position_embeddings": 32768,
|
| 88 |
"max_window_layers": 28,
|
| 89 |
+
"model_type": "qwen2_vl_text",
|
| 90 |
"num_attention_heads": 12,
|
| 91 |
"num_hidden_layers": 28,
|
| 92 |
"num_key_value_heads": 2,
|
| 93 |
"pad_token_id": 151654,
|
| 94 |
+
"quantization_config": {
|
| 95 |
+
"_load_in_4bit": true,
|
| 96 |
+
"_load_in_8bit": false,
|
| 97 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 98 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 99 |
+
"bnb_4bit_quant_type": "nf4",
|
| 100 |
+
"bnb_4bit_use_double_quant": true,
|
| 101 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 102 |
+
"llm_int8_has_fp16_weight": false,
|
| 103 |
+
"llm_int8_skip_modules": [
|
| 104 |
+
"lm_head",
|
| 105 |
+
"multi_modal_projector",
|
| 106 |
+
"merger",
|
| 107 |
+
"modality_projection",
|
| 108 |
+
"model.layers.1.mlp",
|
| 109 |
+
"visual.blocks.29.mlp",
|
| 110 |
+
"visual.blocks.21.mlp",
|
| 111 |
+
"visual.blocks.24.attn",
|
| 112 |
+
"visual.blocks.26.mlp",
|
| 113 |
+
"visual.blocks.23.attn",
|
| 114 |
+
"visual.blocks.18.mlp",
|
| 115 |
+
"visual.blocks.19.mlp",
|
| 116 |
+
"visual.blocks.22.attn",
|
| 117 |
+
"visual.blocks.17.mlp",
|
| 118 |
+
"visual.blocks.25.attn",
|
| 119 |
+
"visual.blocks.24.mlp",
|
| 120 |
+
"visual.blocks.26.attn",
|
| 121 |
+
"visual.blocks.29.attn",
|
| 122 |
+
"visual.blocks.21.attn",
|
| 123 |
+
"visual.blocks.22.mlp",
|
| 124 |
+
"visual.blocks.14.mlp",
|
| 125 |
+
"visual.blocks.23.mlp",
|
| 126 |
+
"visual.blocks.14.attn",
|
| 127 |
+
"visual.merger.mlp",
|
| 128 |
+
"visual.blocks.16.mlp",
|
| 129 |
+
"visual.blocks.20.mlp",
|
| 130 |
+
"visual.blocks.25.mlp",
|
| 131 |
+
"visual.blocks.13.mlp",
|
| 132 |
+
"visual.blocks.18.attn",
|
| 133 |
+
"visual.blocks.20.attn",
|
| 134 |
+
"visual.blocks.17.attn",
|
| 135 |
+
"visual.blocks.15.mlp",
|
| 136 |
+
"visual.blocks.27.attn",
|
| 137 |
+
"visual.blocks.28.attn",
|
| 138 |
+
"visual.blocks.12.mlp",
|
| 139 |
+
"visual.blocks.16.attn",
|
| 140 |
+
"visual.blocks.19.attn",
|
| 141 |
+
"visual.blocks.15.attn",
|
| 142 |
+
"visual.blocks.11.attn",
|
| 143 |
+
"visual.blocks.13.attn",
|
| 144 |
+
"visual.blocks.12.attn",
|
| 145 |
+
"visual.blocks.10.mlp",
|
| 146 |
+
"visual.blocks.11.mlp",
|
| 147 |
+
"visual.blocks.10.attn",
|
| 148 |
+
"visual.blocks.9.mlp",
|
| 149 |
+
"visual.blocks.5.attn",
|
| 150 |
+
"visual.blocks.9.attn",
|
| 151 |
+
"visual.blocks.4.attn",
|
| 152 |
+
"visual.blocks.6.mlp",
|
| 153 |
+
"visual.blocks.7.mlp",
|
| 154 |
+
"visual.blocks.1.attn",
|
| 155 |
+
"visual.blocks.6.attn",
|
| 156 |
+
"visual.blocks.7.attn",
|
| 157 |
+
"visual.blocks.5.mlp",
|
| 158 |
+
"visual.blocks.8.attn",
|
| 159 |
+
"visual.blocks.8.mlp",
|
| 160 |
+
"visual.blocks.4.mlp",
|
| 161 |
+
"visual.blocks.3.attn",
|
| 162 |
+
"visual.blocks.3.mlp",
|
| 163 |
+
"visual.blocks.1.mlp",
|
| 164 |
+
"visual.blocks.2.mlp",
|
| 165 |
+
"visual.blocks.2.attn",
|
| 166 |
+
"visual.blocks.0.attn",
|
| 167 |
+
"visual.blocks.0.mlp",
|
| 168 |
+
"visual.blocks.31.mlp"
|
| 169 |
+
],
|
| 170 |
+
"llm_int8_threshold": 6.0,
|
| 171 |
+
"load_in_4bit": true,
|
| 172 |
+
"load_in_8bit": false,
|
| 173 |
+
"quant_method": "bitsandbytes"
|
| 174 |
+
},
|
| 175 |
"rms_norm_eps": 1e-06,
|
| 176 |
"rope_scaling": {
|
| 177 |
+
"mrope_section": [
|
| 178 |
+
16,
|
| 179 |
+
24,
|
| 180 |
+
24
|
| 181 |
+
],
|
| 182 |
+
"rope_type": "default",
|
| 183 |
+
"type": "default"
|
| 184 |
},
|
| 185 |
"rope_theta": 1000000.0,
|
| 186 |
+
"sliding_window": null,
|
| 187 |
+
"tie_word_embeddings": true,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 188 |
"unsloth_fixed": true,
|
|
|
|
| 189 |
"use_cache": true,
|
| 190 |
"use_sliding_window": false,
|
| 191 |
+
"video_token_id": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 192 |
"vision_end_token_id": 151653,
|
| 193 |
"vision_start_token_id": 151652,
|
| 194 |
"vision_token_id": 151654,
|
| 195 |
"vocab_size": 151936
|
| 196 |
+
},
|
| 197 |
+
"transformers_version": "4.56.2",
|
| 198 |
+
"unsloth_fixed": true,
|
| 199 |
+
"unsloth_version": "2025.9.11",
|
| 200 |
+
"use_cache": true,
|
| 201 |
+
"use_sliding_window": false,
|
| 202 |
+
"video_token_id": 151656,
|
| 203 |
+
"vision_config": {
|
| 204 |
+
"depth": 32,
|
| 205 |
+
"dtype": "float16",
|
| 206 |
+
"embed_dim": 1280,
|
| 207 |
+
"hidden_act": "quick_gelu",
|
| 208 |
+
"hidden_size": 1536,
|
| 209 |
+
"in_channels": 3,
|
| 210 |
+
"in_chans": 3,
|
| 211 |
+
"initializer_range": 0.02,
|
| 212 |
+
"mlp_ratio": 4,
|
| 213 |
+
"model_type": "qwen2_vl",
|
| 214 |
+
"num_heads": 16,
|
| 215 |
+
"patch_size": 14,
|
| 216 |
+
"spatial_merge_size": 2,
|
| 217 |
+
"spatial_patch_size": 14,
|
| 218 |
+
"temporal_patch_size": 2
|
| 219 |
+
},
|
| 220 |
+
"vision_end_token_id": 151653,
|
| 221 |
+
"vision_start_token_id": 151652,
|
| 222 |
+
"vision_token_id": 151654,
|
| 223 |
+
"vocab_size": 151936
|
| 224 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": 151645,
|
| 4 |
+
"max_length": 32768,
|
| 5 |
+
"pad_token_id": 151654,
|
| 6 |
+
"transformers_version": "4.56.2"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a177667940f09e5cec23b851d06274c83f5dd7362997622aba3c8fca0c12a731
|
| 3 |
+
size 1540409944
|