| { |
| "architectures": [ |
| "Qwen4ExpForConditionalGeneration" |
| ], |
| "image_token_id": 248056, |
| "language_model_only": false, |
| "mlx_lm_extra_tensors": { |
| "mtp_file": "mtp.safetensors", |
| "mtp_format": "torch-layout-fp16-v1", |
| "mtp_reason": "MTP is retained outside the trunk because the referenced MLX qwen4_exp model has no MTP module; module-blind quantization would not be load-safe.", |
| "vision_file": "vision.safetensors", |
| "vision_format": "torch-layout-source-dtype-v1", |
| "vision_reason": "The referenced MLX qwen4_exp model is text-only; the complete vision tower is retained by key, dtype, and shape in a sidecar." |
| }, |
| "model_type": "qwen4_exp", |
| "quantization": { |
| "bits": 4, |
| "group_size": 64, |
| "mode": "affine", |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_0": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_1": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_10": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_100": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_101": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_102": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_103": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_104": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_105": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_106": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_107": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_108": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_109": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_11": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_110": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_111": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_112": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_113": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_114": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_115": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_116": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_117": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_118": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_119": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_12": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_120": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_121": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_122": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_123": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_124": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_125": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_126": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_127": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_13": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_14": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_15": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_16": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_17": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_18": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_19": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_2": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_20": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_21": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_22": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_23": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_24": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_25": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_26": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_27": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_28": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_29": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_3": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_30": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_31": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_32": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_33": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_34": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_35": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_36": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_37": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_38": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_39": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_4": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_40": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_41": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_42": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_43": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_44": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_45": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_46": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_47": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_48": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_49": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_5": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_50": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_51": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_52": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_53": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_54": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_55": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_56": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_57": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_58": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_59": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_6": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_60": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_61": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_62": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_63": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_64": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_65": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_66": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_67": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_68": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_69": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_7": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_70": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_71": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_72": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_73": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_74": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_75": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_76": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_77": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_78": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_79": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_8": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_80": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_81": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_82": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_83": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_84": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_85": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_86": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_87": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_88": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_89": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_9": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_90": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_91": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_92": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_93": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_94": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_95": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_96": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_97": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_98": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_99": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| } |
| }, |
| "quantization_config": { |
| "bits": 4, |
| "group_size": 64, |
| "mode": "affine", |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_0": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_1": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_10": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_100": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_101": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_102": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_103": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_104": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_105": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_106": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_107": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_108": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_109": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_11": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_110": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_111": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_112": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_113": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_114": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_115": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_116": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_117": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_118": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_119": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_12": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_120": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_121": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_122": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_123": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_124": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_125": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_126": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_127": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_13": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_14": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_15": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_16": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_17": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_18": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_19": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_2": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_20": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_21": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_22": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_23": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_24": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_25": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_26": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_27": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_28": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_29": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_3": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_30": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_31": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_32": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_33": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_34": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_35": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_36": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_37": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_38": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_39": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_4": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_40": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_41": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_42": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_43": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_44": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_45": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_46": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_47": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_48": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_49": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_5": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_50": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_51": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_52": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_53": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_54": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_55": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_56": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_57": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_58": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_59": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_6": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_60": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_61": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_62": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_63": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_64": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_65": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_66": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_67": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_68": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_69": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_7": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_70": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_71": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_72": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_73": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_74": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_75": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_76": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_77": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_78": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_79": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_8": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_80": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_81": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_82": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_83": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_84": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_85": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_86": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_87": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_88": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_89": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_9": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_90": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_91": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_92": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_93": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_94": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_95": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_96": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_97": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_98": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| }, |
| "model.layers.1.ple.ple_embedding.ngram_embedding.shard_99": { |
| "bits": 4, |
| "group_size": 32, |
| "mode": "affine" |
| } |
| }, |
| "text_config": { |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 248044, |
| "dtype": "bfloat16", |
| "eos_token_id": 248044, |
| "full_attention_interval": 4, |
| "hc_count": 4, |
| "hc_lowrank": 320, |
| "head_dim": 256, |
| "heads_per_ngram": 8, |
| "hidden_act": "silu", |
| "hidden_size": 2560, |
| "indexer_budget": 2048, |
| "indexer_compress_ratio": 4, |
| "indexer_head_dim": 128, |
| "indexer_kv_heads": 1, |
| "indexer_n_heads": 4, |
| "initializer_range": 0.02, |
| "layer_types": [ |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention" |
| ], |
| "linear_conv_kernel_dim": 4, |
| "linear_key_head_dim": 128, |
| "linear_num_key_heads": 16, |
| "linear_num_value_heads": 48, |
| "linear_value_head_dim": 128, |
| "make_ngram_vocab_size_divisible_by": 128, |
| "mamba_ssm_dtype": "float32", |
| "max_position_embeddings": 262144, |
| "model_type": "qwen4_exp_text", |
| "moe_intermediate_size": 640, |
| "mtp": { |
| "hybrid": true, |
| "layer_types": [ |
| "full_attention" |
| ], |
| "mtp_use_hidden_state_from_layer": null, |
| "num_hidden_layers": 1, |
| "rope_theta": 10000000 |
| }, |
| "mtp_num_hidden_layers": 1, |
| "mtp_use_dedicated_embeddings": false, |
| "ngram_size": 3, |
| "ngram_vocab_size_base": 20000000, |
| "num_attention_heads": 24, |
| "num_experts": 512, |
| "num_experts_per_tok": 10, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 2, |
| "output_gate_type": "sigmoid", |
| "output_router_logits": false, |
| "pad_token_id": null, |
| "partial_rotary_factor": 0.25, |
| "ple_conv_kernel_size": 4, |
| "ple_embed_dim": 2560, |
| "ple_layer_ids": [ |
| 2 |
| ], |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "mrope_interleaved": true, |
| "mrope_section": [ |
| 11, |
| 11, |
| 10 |
| ], |
| "partial_rotary_factor": 0.25, |
| "rope_theta": 10000000, |
| "rope_type": "default" |
| }, |
| "router_aux_loss_coef": 0.001, |
| "shared_expert_intermediate_size": 640, |
| "split_ngram_parts": 128, |
| "tie_word_embeddings": false, |
| "use_cache": true, |
| "vocab_size": 248320 |
| }, |
| "tie_word_embeddings": false, |
| "transformers_version": "5.8.0.dev0", |
| "video_token_id": 248057, |
| "vision_config": { |
| "deepstack_visual_indexes": [], |
| "depth": 27, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_size": 1152, |
| "in_channels": 3, |
| "initializer_range": 0.02, |
| "intermediate_size": 4304, |
| "model_type": "qwen4_exp", |
| "num_heads": 16, |
| "num_position_embeddings": 2304, |
| "out_hidden_size": 2560, |
| "patch_size": 16, |
| "spatial_merge_size": 2, |
| "temporal_patch_size": 2 |
| }, |
| "vision_end_token_id": 248054, |
| "vision_start_token_id": 248053 |
| } |
|
|
