BellLabs/DAS_spectrogram_Qwen2.5-VL-3B-Instruct
08
1{2 "architectures": [3 "Qwen2_5_VLForConditionalGeneration"4 ],5 "attention_dropout": 0.0,6 "bos_token_id": 151643,7 "eos_token_id": 151645,8 "hidden_act": "silu",9 "hidden_size": 2048,10 "image_token_id": 151655,11 "initializer_range": 0.02,12 "intermediate_size": 11008,13 "max_position_embeddings": 128000,14 "max_window_layers": 70,15 "model_type": "qwen2_5_vl",16 "num_attention_heads": 16,17 "num_hidden_layers": 36,18 "num_key_value_heads": 2,19 "rms_norm_eps": 1e-06,20 "rope_scaling": {21 "mrope_section": [22 16,23 24,24 2425 ],26 "rope_type": "default",27 "type": "default"28 },29 "rope_theta": 1000000.0,30 "sliding_window": 32768,31 "text_config": {32 "architectures": [33 "Qwen2_5_VLForConditionalGeneration"34 ],35 "attention_dropout": 0.0,36 "bos_token_id": 151643,37 "eos_token_id": 151645,38 "hidden_act": "silu",39 "hidden_size": 2048,40 "image_token_id": null,41 "initializer_range": 0.02,42 "intermediate_size": 11008,43 "layer_types": [44 "full_attention",45 "full_attention",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 "full_attention",75 "full_attention",76 "full_attention",77 "full_attention",78 "full_attention",79 "full_attention"80 ],81 "max_position_embeddings": 128000,82 "max_window_layers": 70,83 "model_type": "qwen2_5_vl_text",84 "num_attention_heads": 16,85 "num_hidden_layers": 36,86 "num_key_value_heads": 2,87 "rms_norm_eps": 1e-06,88 "rope_scaling": {89 "mrope_section": [90 16,91 24,92 2493 ],94 "rope_type": "default",95 "type": "default"96 },97 "rope_theta": 1000000.0,98 "sliding_window": null,99 "tie_word_embeddings": true,100 "torch_dtype": "bfloat16",101 "use_cache": true,102 "use_sliding_window": false,103 "video_token_id": null,104 "vision_end_token_id": 151653,105 "vision_start_token_id": 151652,106 "vision_token_id": 151654,107 "vocab_size": 151936108 },109 "torch_dtype": "bfloat16",110 "transformers_version": "4.53.1",111 "use_cache": true,112 "use_sliding_window": false,113 "video_token_id": 151656,114 "vision_config": {115 "depth": 32,116 "fullatt_block_indexes": [117 7,118 15,119 23,120 31121 ],122 "hidden_act": "silu",123 "hidden_size": 1280,124 "in_channels": 3,125 "in_chans": 3,126 "initializer_range": 0.02,127 "intermediate_size": 3420,128 "model_type": "qwen2_5_vl",129 "num_heads": 16,130 "out_hidden_size": 2048,131 "patch_size": 14,132 "spatial_merge_size": 2,133 "spatial_patch_size": 14,134 "temporal_patch_size": 2,135 "tokens_per_second": 2,136 "torch_dtype": "bfloat16",137 "window_size": 112138 },139 "vision_end_token_id": 151653,140 "vision_start_token_id": 151652,141 "vision_token_id": 151654,142 "vocab_size": 151936143}144 