devJy/space-voice-label-detect-beta

1{2    "architectures": [3        "Qwen2_5_VLModel"4    ],5    "attention_dropout": 0.0,6    "eos_token_id": 151645,7    "hidden_act": "silu",8    "hidden_size": 2048,9    "image_token_id": 151655,10    "initializer_range": 0.02,11    "intermediate_size": 11008,12    "max_position_embeddings": 32768,13    "max_window_layers": 70,14    "model_type": "qwen2_5_vl",15    "num_attention_heads": 16,16    "num_hidden_layers": 36,17    "num_key_value_heads": 2,18    "pad_token_id": 151654,19    "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    "tie_word_embeddings": true,32    "torch_dtype": "bfloat16",33    "transformers_version": "4.50.3",34    "unsloth_fixed": true,35    "unsloth_version": "2025.3.19",36    "use_cache": true,37    "use_sliding_window": false,38    "video_token_id": 151656,39    "vision_config": {40        "depth": 32,41        "fullatt_block_indexes": [42            7,43            15,44            23,45            3146        ],47        "hidden_act": "silu",48        "hidden_size": 1280,49        "in_channels": 3,50        "in_chans": 3,51        "intermediate_size": 3420,52        "model_type": "qwen2_5_vl",53        "num_heads": 16,54        "out_hidden_size": 2048,55        "patch_size": 14,56        "spatial_merge_size": 2,57        "spatial_patch_size": 14,58        "temporal_patch_size": 2,59        "tokens_per_second": 2,60        "torch_dtype": "bfloat16",61        "window_size": 11262    },63    "vision_end_token_id": 151653,64    "vision_start_token_id": 151652,65    "vision_token_id": 151654,66    "vocab_size": 15193667}