CoolFace
Modelpublic

nathanles/RustyBender

sourceHugging Faceapache-2.0updated 15d agoView on Hugging Face
0likes23downloads
config.json166 linesDownload Raw Back to root
1{2    "architectures": [3        "Gemma4UnifiedForConditionalGeneration"4    ],5    "audio_config": {6        "_name_or_path": "",7        "architectures": null,8        "audio_embed_dim": 640,9        "chunk_size_feed_forward": 0,10        "torch_dtype": null,11        "id2label": {12            "0": "LABEL_0",13            "1": "LABEL_1"14        },15        "initializer_range": 0.02,16        "is_encoder_decoder": false,17        "label2id": {18            "LABEL_0": 0,19            "LABEL_1": 120        },21        "model_type": "gemma4_unified_audio",22        "output_attentions": false,23        "output_hidden_states": false,24        "problem_type": null,25        "return_dict": true,26        "rms_norm_eps": 1e-0627    },28    "audio_token_id": 258881,29    "boa_token_id": 256000,30    "boi_token_id": 255999,31    "torch_dtype": "bfloat16",32    "eoa_token_index": 258883,33    "eoi_token_id": 258882,34    "eos_token_id": 106,35    "image_token_id": 258880,36    "initializer_range": 0.02,37    "model_type": "gemma4_unified",38    "pad_token_id": 0,39    "text_config": {40        "attention_bias": false,41        "attention_dropout": 0.0,42        "attention_k_eq_v": true,43        "bos_token_id": 2,44        "enable_moe_block": false,45        "eos_token_id": 1,46        "final_logit_softcapping": 30.0,47        "global_head_dim": 512,48        "head_dim": 256,49        "hidden_activation": "gelu_pytorch_tanh",50        "hidden_size": 3840,51        "hidden_size_per_layer_input": 0,52        "initializer_range": 0.02,53        "intermediate_size": 15360,54        "layer_types": [55            "sliding_attention",56            "sliding_attention",57            "sliding_attention",58            "sliding_attention",59            "sliding_attention",60            "full_attention",61            "sliding_attention",62            "sliding_attention",63            "sliding_attention",64            "sliding_attention",65            "sliding_attention",66            "full_attention",67            "sliding_attention",68            "sliding_attention",69            "sliding_attention",70            "sliding_attention",71            "sliding_attention",72            "full_attention",73            "sliding_attention",74            "sliding_attention",75            "sliding_attention",76            "sliding_attention",77            "sliding_attention",78            "full_attention",79            "sliding_attention",80            "sliding_attention",81            "sliding_attention",82            "sliding_attention",83            "sliding_attention",84            "full_attention",85            "sliding_attention",86            "sliding_attention",87            "sliding_attention",88            "sliding_attention",89            "sliding_attention",90            "full_attention",91            "sliding_attention",92            "sliding_attention",93            "sliding_attention",94            "sliding_attention",95            "sliding_attention",96            "full_attention",97            "sliding_attention",98            "sliding_attention",99            "sliding_attention",100            "sliding_attention",101            "sliding_attention",102            "full_attention"103        ],104        "max_position_embeddings": 262144,105        "model_type": "gemma4_unified_text",106        "moe_intermediate_size": null,107        "num_attention_heads": 16,108        "num_experts": null,109        "num_global_key_value_heads": 1,110        "num_hidden_layers": 48,111        "num_key_value_heads": 8,112        "num_kv_shared_layers": 0,113        "pad_token_id": 0,114        "rms_norm_eps": 1e-06,115        "rope_parameters": {116            "full_attention": {117                "partial_rotary_factor": 0.25,118                "rope_theta": 1000000.0,119                "rope_type": "proportional"120            },121            "sliding_attention": {122                "rope_theta": 10000.0,123                "rope_type": "default"124            }125        },126        "sliding_window": 1024,127        "tie_word_embeddings": true,128        "top_k_experts": null,129        "use_bidirectional_attention": "vision",130        "use_cache": true,131        "use_double_wide_mlp": false,132        "vocab_size": 262144,133        "vocab_size_per_layer_input": 262144134    },135    "tie_word_embeddings": true,136    "unsloth_fixed": true,137    "video_token_id": 258884,138    "vision_config": {139        "_name_or_path": "",140        "architectures": null,141        "chunk_size_feed_forward": 0,142        "torch_dtype": null,143        "id2label": {144            "0": "LABEL_0",145            "1": "LABEL_1"146        },147        "initializer_range": 0.02,148        "is_encoder_decoder": false,149        "label2id": {150            "LABEL_0": 0,151            "LABEL_1": 1152        },153        "mm_embed_dim": 3840,154        "mm_posemb_size": 1120,155        "model_type": "gemma4_unified_vision",156        "num_soft_tokens": 280,157        "output_attentions": false,158        "output_hidden_states": false,159        "output_proj_dims": 3840,160        "patch_size": 16,161        "pooling_kernel_size": 3,162        "problem_type": null,163        "return_dict": true,164        "rms_norm_eps": 1e-06165    }166}