Download config.json from tencent/WeMM-Video-9B-FP8: direct link, hf CLI and curl.
- Browser
- Download file 52.2 kB
-
https://huggingface.co/tencent/WeMM-Video-9B-FP8/resolve/main/config.json
- Command line
-
hf download hf://tencent/WeMM-Video-9B-FP8/config.json
-
curl -L -o config.json https://huggingface.co/tencent/WeMM-Video-9B-FP8/resolve/main/config.json
52.2 kB
{
"architectures": [
"WeMMVideoForConditionalGeneration"
],
"audio_config": {
"activation_dropout": 0.0,
"activation_function": "gelu",
"apply_spec_augment": false,
"attention_dropout": 0.0,
"bos_token_id": 50257,
"classifier_proj_size": 256,
"d_model": 1280,
"decoder_attention_heads": 20,
"decoder_ffn_dim": 5120,
"decoder_layerdrop": 0.0,
"decoder_layers": 32,
"decoder_start_token_id": 50258,
"dropout": 0.0,
"dtype": "bfloat16",
"encoder_attention_heads": 20,
"encoder_ffn_dim": 5120,
"encoder_layerdrop": 0.0,
"encoder_layers": 32,
"eos_token_id": 50257,
"init_std": 0.02,
"mask_feature_length": 10,
"mask_feature_min_masks": 0,
"mask_feature_prob": 0.0,
"mask_time_length": 10,
"mask_time_min_masks": 2,
"mask_time_prob": 0.05,
"max_source_positions": 1500,
"max_target_positions": 448,
"median_filter_width": 7,
"model_type": "whisper",
"num_hidden_layers": 32,
"num_mel_bins": 128,
"pad_token_id": 50256,
"qformer_config": {
"add_cross_attention": true,
"architectures": null,
"attention_probs_dropout_prob": 0.1,
"chunk_size_feed_forward": 0,
"classifier_dropout": null,
"cross_attention_freq": 1,
"dtype": null,
"encoder_width": 1280,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 4104,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"initializer_range": 0.02,
"intermediate_size": 12288,
"is_encoder_decoder": false,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"layer_norm_eps": 1e-12,
"max_position_embeddings": 512,
"model_type": "bert",
"num_attention_heads": 12,
"num_hidden_layers": 2,
"out_hidden_size": 4096,
"output_attentions": false,
"output_hidden_states": false,
"pad_token_id": 0,
"position_embedding_type": "absolute",
"problem_type": null,
"query_length": 1,
"return_dict": true,
"type_vocab_size": 2,
"use_cache": true,
"vocab_size": 30522
},
"scale_embedding": false,
"second_per_window": 0.1,
"second_stride": 0.1,
"tie_word_embeddings": true,
"use_cache": true,
"use_weighted_layer_sum": false,
"vocab_size": 51866
},
"audio_token_id": 248076,
"dtype": "bfloat16",
"image_token_id": 248056,
"model_type": "qwen3_5",
"text_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"attn_output_gate": true,
"bos_token_id": null,
"dtype": "bfloat16",
"eos_token_id": 248044,
"full_attention_interval": 4,
"head_dim": 256,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 12288,
"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_conv_kernel_dim": 4,
"linear_key_head_dim": 128,
"linear_num_key_heads": 16,
"linear_num_value_heads": 32,
"linear_value_head_dim": 128,
"mamba_ssm_dtype": "float32",
"max_position_embeddings": 262144,
"mlp_only_layers": [],
"model_type": "qwen3_5_text",
"mtp_num_hidden_layers": 1,
"mtp_use_dedicated_embeddings": false,
"num_attention_heads": 16,
"num_hidden_layers": 32,
"num_key_value_heads": 4,
"pad_token_id": null,
"partial_rotary_factor": 0.25,
"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"
},
"tie_word_embeddings": false,
"use_cache": true,
"vocab_size": 248320
},
"tie_word_embeddings": false,
"transformers_version": "5.2.0",
"video_token_id": 248057,
"vision_config": {
"deepstack_visual_indexes": [],
"depth": 27,
"dtype": "bfloat16",
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 1152,
"in_channels": 3,
"initializer_range": 0.02,
"intermediate_size": 4304,
"model_type": "qwen3_5",
"num_heads": 16,
"num_position_embeddings": 2304,
"out_hidden_size": 4096,
"patch_size": 16,
"spatial_merge_size": 2,
"temporal_patch_size": 2
},
"vision_end_token_id": 248054,
"vision_start_token_id": 248053,
"auto_map": {
"AutoConfig": "configuration_wemm.WeMMQwen3_5Config",
"AutoModel": "modeling_wemm.WeMMVideoForConditionalGeneration",
"AutoModelForImageTextToText": "modeling_wemm.WeMMVideoForConditionalGeneration",
"AutoProcessor": "processing_wemm.WeMMProcessor"
},
"quantization_config": {
"quant_method": "fp8",
"activation_scheme": "dynamic",
"weight_per_tensor": false,
"act_per_tensor": false,
"weight_block_size": [
128,
128
],
"modules_to_not_convert": [
"lm_head",
"model.audio",
"model.audio.audio_proj",
"model.audio.audio_proj.bias",
"model.audio.conv1",
"model.audio.conv1.bias",
"model.audio.conv2",
"model.audio.conv2.bias",
"model.audio.embed_positions",
"model.audio.layer_norm",
"model.audio.layer_norm.bias",
"model.audio.layers.0.fc1",
"model.audio.layers.0.fc1.bias",
"model.audio.layers.0.fc2",
"model.audio.layers.0.fc2.bias",
"model.audio.layers.0.final_layer_norm",
"model.audio.layers.0.final_layer_norm.bias",
"model.audio.layers.0.self_attn.k_proj",
"model.audio.layers.0.self_attn.out_proj",
"model.audio.layers.0.self_attn.out_proj.bias",
"model.audio.layers.0.self_attn.q_proj",
"model.audio.layers.0.self_attn.q_proj.bias",
"model.audio.layers.0.self_attn.v_proj",
"model.audio.layers.0.self_attn.v_proj.bias",
"model.audio.layers.0.self_attn_layer_norm",
"model.audio.layers.0.self_attn_layer_norm.bias",
"model.audio.layers.1.fc1",
"model.audio.layers.1.fc1.bias",
"model.audio.layers.1.fc2",
"model.audio.layers.1.fc2.bias",
"model.audio.layers.1.final_layer_norm",
"model.audio.layers.1.final_layer_norm.bias",
"model.audio.layers.1.self_attn.k_proj",
"model.audio.layers.1.self_attn.out_proj",
"model.audio.layers.1.self_attn.out_proj.bias",
"model.audio.layers.1.self_attn.q_proj",
"model.audio.layers.1.self_attn.q_proj.bias",
"model.audio.layers.1.self_attn.v_proj",
"model.audio.layers.1.self_attn.v_proj.bias",
"model.audio.layers.1.self_attn_layer_norm",
"model.audio.layers.1.self_attn_layer_norm.bias",
"model.audio.layers.10.fc1",
"model.audio.layers.10.fc1.bias",
"model.audio.layers.10.fc2",
"model.audio.layers.10.fc2.bias",
"model.audio.layers.10.final_layer_norm",
"model.audio.layers.10.final_layer_norm.bias",
"model.audio.layers.10.self_attn.k_proj",
"model.audio.layers.10.self_attn.out_proj",
"model.audio.layers.10.self_attn.out_proj.bias",
"model.audio.layers.10.self_attn.q_proj",
"model.audio.layers.10.self_attn.q_proj.bias",
"model.audio.layers.10.self_attn.v_proj",
"model.audio.layers.10.self_attn.v_proj.bias",
"model.audio.layers.10.self_attn_layer_norm",
"model.audio.layers.10.self_attn_layer_norm.bias",
"model.audio.layers.11.fc1",
"model.audio.layers.11.fc1.bias",
"model.audio.layers.11.fc2",
"model.audio.layers.11.fc2.bias",
"model.audio.layers.11.final_layer_norm",
"model.audio.layers.11.final_layer_norm.bias",
"model.audio.layers.11.self_attn.k_proj",
"model.audio.layers.11.self_attn.out_proj",
"model.audio.layers.11.self_attn.out_proj.bias",
"model.audio.layers.11.self_attn.q_proj",
"model.audio.layers.11.self_attn.q_proj.bias",
"model.audio.layers.11.self_attn.v_proj",
"model.audio.layers.11.self_attn.v_proj.bias",
"model.audio.layers.11.self_attn_layer_norm",
"model.audio.layers.11.self_attn_layer_norm.bias",
"model.audio.layers.12.fc1",
"model.audio.layers.12.fc1.bias",
"model.audio.layers.12.fc2",
"model.audio.layers.12.fc2.bias",
"model.audio.layers.12.final_layer_norm",
"model.audio.layers.12.final_layer_norm.bias",
"model.audio.layers.12.self_attn.k_proj",
"model.audio.layers.12.self_attn.out_proj",
"model.audio.layers.12.self_attn.out_proj.bias",
"model.audio.layers.12.self_attn.q_proj",
"model.audio.layers.12.self_attn.q_proj.bias",
"model.audio.layers.12.self_attn.v_proj",
"model.audio.layers.12.self_attn.v_proj.bias",
"model.audio.layers.12.self_attn_layer_norm",
"model.audio.layers.12.self_attn_layer_norm.bias",
"model.audio.layers.13.fc1",
"model.audio.layers.13.fc1.bias",
"model.audio.layers.13.fc2",
"model.audio.layers.13.fc2.bias",
"model.audio.layers.13.final_layer_norm",
"model.audio.layers.13.final_layer_norm.bias",
"model.audio.layers.13.self_attn.k_proj",
"model.audio.layers.13.self_attn.out_proj",
"model.audio.layers.13.self_attn.out_proj.bias",
"model.audio.layers.13.self_attn.q_proj",
"model.audio.layers.13.self_attn.q_proj.bias",
"model.audio.layers.13.self_attn.v_proj",
"model.audio.layers.13.self_attn.v_proj.bias",
"model.audio.layers.13.self_attn_layer_norm",
"model.audio.layers.13.self_attn_layer_norm.bias",
"model.audio.layers.14.fc1",
"model.audio.layers.14.fc1.bias",
"model.audio.layers.14.fc2",
"model.audio.layers.14.fc2.bias",
"model.audio.layers.14.final_layer_norm",
"model.audio.layers.14.final_layer_norm.bias",
"model.audio.layers.14.self_attn.k_proj",
"model.audio.layers.14.self_attn.out_proj",
"model.audio.layers.14.self_attn.out_proj.bias",
"model.audio.layers.14.self_attn.q_proj",
"model.audio.layers.14.self_attn.q_proj.bias",
"model.audio.layers.14.self_attn.v_proj",
"model.audio.layers.14.self_attn.v_proj.bias",
"model.audio.layers.14.self_attn_layer_norm",
"model.audio.layers.14.self_attn_layer_norm.bias",
"model.audio.layers.15.fc1",
"model.audio.layers.15.fc1.bias",
"model.audio.layers.15.fc2",
"model.audio.layers.15.fc2.bias",
"model.audio.layers.15.final_layer_norm",
"model.audio.layers.15.final_layer_norm.bias",
"model.audio.layers.15.self_attn.k_proj",
"model.audio.layers.15.self_attn.out_proj",
"model.audio.layers.15.self_attn.out_proj.bias",
"model.audio.layers.15.self_attn.q_proj",
"model.audio.layers.15.self_attn.q_proj.bias",
"model.audio.layers.15.self_attn.v_proj",
"model.audio.layers.15.self_attn.v_proj.bias",
"model.audio.layers.15.self_attn_layer_norm",
"model.audio.layers.15.self_attn_layer_norm.bias",
"model.audio.layers.16.fc1",
"model.audio.layers.16.fc1.bias",
"model.audio.layers.16.fc2",
"model.audio.layers.16.fc2.bias",
"model.audio.layers.16.final_layer_norm",
"model.audio.layers.16.final_layer_norm.bias",
"model.audio.layers.16.self_attn.k_proj",
"model.audio.layers.16.self_attn.out_proj",
"model.audio.layers.16.self_attn.out_proj.bias",
"model.audio.layers.16.self_attn.q_proj",
"model.audio.layers.16.self_attn.q_proj.bias",
"model.audio.layers.16.self_attn.v_proj",
"model.audio.layers.16.self_attn.v_proj.bias",
"model.audio.layers.16.self_attn_layer_norm",
"model.audio.layers.16.self_attn_layer_norm.bias",
"model.audio.layers.17.fc1",
"model.audio.layers.17.fc1.bias",
"model.audio.layers.17.fc2",
"model.audio.layers.17.fc2.bias",
"model.audio.layers.17.final_layer_norm",
"model.audio.layers.17.final_layer_norm.bias",
"model.audio.layers.17.self_attn.k_proj",
"model.audio.layers.17.self_attn.out_proj",
"model.audio.layers.17.self_attn.out_proj.bias",
"model.audio.layers.17.self_attn.q_proj",
"model.audio.layers.17.self_attn.q_proj.bias",
"model.audio.layers.17.self_attn.v_proj",
"model.audio.layers.17.self_attn.v_proj.bias",
"model.audio.layers.17.self_attn_layer_norm",
"model.audio.layers.17.self_attn_layer_norm.bias",
"model.audio.layers.18.fc1",
"model.audio.layers.18.fc1.bias",
"model.audio.layers.18.fc2",
"model.audio.layers.18.fc2.bias",
"model.audio.layers.18.final_layer_norm",
"model.audio.layers.18.final_layer_norm.bias",
"model.audio.layers.18.self_attn.k_proj",
"model.audio.layers.18.self_attn.out_proj",
"model.audio.layers.18.self_attn.out_proj.bias",
"model.audio.layers.18.self_attn.q_proj",
"model.audio.layers.18.self_attn.q_proj.bias",
"model.audio.layers.18.self_attn.v_proj",
"model.audio.layers.18.self_attn.v_proj.bias",
"model.audio.layers.18.self_attn_layer_norm",
"model.audio.layers.18.self_attn_layer_norm.bias",
"model.audio.layers.19.fc1",
"model.audio.layers.19.fc1.bias",
"model.audio.layers.19.fc2",
"model.audio.layers.19.fc2.bias",
"model.audio.layers.19.final_layer_norm",
"model.audio.layers.19.final_layer_norm.bias",
"model.audio.layers.19.self_attn.k_proj",
"model.audio.layers.19.self_attn.out_proj",
"model.audio.layers.19.self_attn.out_proj.bias",
"model.audio.layers.19.self_attn.q_proj",
"model.audio.layers.19.self_attn.q_proj.bias",
"model.audio.layers.19.self_attn.v_proj",
"model.audio.layers.19.self_attn.v_proj.bias",
"model.audio.layers.19.self_attn_layer_norm",
"model.audio.layers.19.self_attn_layer_norm.bias",
"model.audio.layers.2.fc1",
"model.audio.layers.2.fc1.bias",
"model.audio.layers.2.fc2",
"model.audio.layers.2.fc2.bias",
"model.audio.layers.2.final_layer_norm",
"model.audio.layers.2.final_layer_norm.bias",
"model.audio.layers.2.self_attn.k_proj",
"model.audio.layers.2.self_attn.out_proj",
"model.audio.layers.2.self_attn.out_proj.bias",
"model.audio.layers.2.self_attn.q_proj",
"model.audio.layers.2.self_attn.q_proj.bias",
"model.audio.layers.2.self_attn.v_proj",
"model.audio.layers.2.self_attn.v_proj.bias",
"model.audio.layers.2.self_attn_layer_norm",
"model.audio.layers.2.self_attn_layer_norm.bias",
"model.audio.layers.20.fc1",
"model.audio.layers.20.fc1.bias",
"model.audio.layers.20.fc2",
"model.audio.layers.20.fc2.bias",
"model.audio.layers.20.final_layer_norm",
"model.audio.layers.20.final_layer_norm.bias",
"model.audio.layers.20.self_attn.k_proj",
"model.audio.layers.20.self_attn.out_proj",
"model.audio.layers.20.self_attn.out_proj.bias",
"model.audio.layers.20.self_attn.q_proj",
"model.audio.layers.20.self_attn.q_proj.bias",
"model.audio.layers.20.self_attn.v_proj",
"model.audio.layers.20.self_attn.v_proj.bias",
"model.audio.layers.20.self_attn_layer_norm",
"model.audio.layers.20.self_attn_layer_norm.bias",
"model.audio.layers.21.fc1",
"model.audio.layers.21.fc1.bias",
"model.audio.layers.21.fc2",
"model.audio.layers.21.fc2.bias",
"model.audio.layers.21.final_layer_norm",
"model.audio.layers.21.final_layer_norm.bias",
"model.audio.layers.21.self_attn.k_proj",
"model.audio.layers.21.self_attn.out_proj",
"model.audio.layers.21.self_attn.out_proj.bias",
"model.audio.layers.21.self_attn.q_proj",
"model.audio.layers.21.self_attn.q_proj.bias",
"model.audio.layers.21.self_attn.v_proj",
"model.audio.layers.21.self_attn.v_proj.bias",
"model.audio.layers.21.self_attn_layer_norm",
"model.audio.layers.21.self_attn_layer_norm.bias",
"model.audio.layers.22.fc1",
"model.audio.layers.22.fc1.bias",
"model.audio.layers.22.fc2",
"model.audio.layers.22.fc2.bias",
"model.audio.layers.22.final_layer_norm",
"model.audio.layers.22.final_layer_norm.bias",
"model.audio.layers.22.self_attn.k_proj",
"model.audio.layers.22.self_attn.out_proj",
"model.audio.layers.22.self_attn.out_proj.bias",
"model.audio.layers.22.self_attn.q_proj",
"model.audio.layers.22.self_attn.q_proj.bias",
"model.audio.layers.22.self_attn.v_proj",
"model.audio.layers.22.self_attn.v_proj.bias",
"model.audio.layers.22.self_attn_layer_norm",
"model.audio.layers.22.self_attn_layer_norm.bias",
"model.audio.layers.23.fc1",
"model.audio.layers.23.fc1.bias",
"model.audio.layers.23.fc2",
"model.audio.layers.23.fc2.bias",
"model.audio.layers.23.final_layer_norm",
"model.audio.layers.23.final_layer_norm.bias",
"model.audio.layers.23.self_attn.k_proj",
"model.audio.layers.23.self_attn.out_proj",
"model.audio.layers.23.self_attn.out_proj.bias",
"model.audio.layers.23.self_attn.q_proj",
"model.audio.layers.23.self_attn.q_proj.bias",
"model.audio.layers.23.self_attn.v_proj",
"model.audio.layers.23.self_attn.v_proj.bias",
"model.audio.layers.23.self_attn_layer_norm",
"model.audio.layers.23.self_attn_layer_norm.bias",
"model.audio.layers.24.fc1",
"model.audio.layers.24.fc1.bias",
"model.audio.layers.24.fc2",
"model.audio.layers.24.fc2.bias",
"model.audio.layers.24.final_layer_norm",
"model.audio.layers.24.final_layer_norm.bias",
"model.audio.layers.24.self_attn.k_proj",
"model.audio.layers.24.self_attn.out_proj",
"model.audio.layers.24.self_attn.out_proj.bias",
"model.audio.layers.24.self_attn.q_proj",
"model.audio.layers.24.self_attn.q_proj.bias",
"model.audio.layers.24.self_attn.v_proj",
"model.audio.layers.24.self_attn.v_proj.bias",
"model.audio.layers.24.self_attn_layer_norm",
"model.audio.layers.24.self_attn_layer_norm.bias",
"model.audio.layers.25.fc1",
"model.audio.layers.25.fc1.bias",
"model.audio.layers.25.fc2",
"model.audio.layers.25.fc2.bias",
"model.audio.layers.25.final_layer_norm",
"model.audio.layers.25.final_layer_norm.bias",
"model.audio.layers.25.self_attn.k_proj",
"model.audio.layers.25.self_attn.out_proj",
"model.audio.layers.25.self_attn.out_proj.bias",
"model.audio.layers.25.self_attn.q_proj",
"model.audio.layers.25.self_attn.q_proj.bias",
"model.audio.layers.25.self_attn.v_proj",
"model.audio.layers.25.self_attn.v_proj.bias",
"model.audio.layers.25.self_attn_layer_norm",
"model.audio.layers.25.self_attn_layer_norm.bias",
"model.audio.layers.26.fc1",
"model.audio.layers.26.fc1.bias",
"model.audio.layers.26.fc2",
"model.audio.layers.26.fc2.bias",
"model.audio.layers.26.final_layer_norm",
"model.audio.layers.26.final_layer_norm.bias",
"model.audio.layers.26.self_attn.k_proj",
"model.audio.layers.26.self_attn.out_proj",
"model.audio.layers.26.self_attn.out_proj.bias",
"model.audio.layers.26.self_attn.q_proj",
"model.audio.layers.26.self_attn.q_proj.bias",
"model.audio.layers.26.self_attn.v_proj",
"model.audio.layers.26.self_attn.v_proj.bias",
"model.audio.layers.26.self_attn_layer_norm",
"model.audio.layers.26.self_attn_layer_norm.bias",
"model.audio.layers.27.fc1",
"model.audio.layers.27.fc1.bias",
"model.audio.layers.27.fc2",
"model.audio.layers.27.fc2.bias",
"model.audio.layers.27.final_layer_norm",
"model.audio.layers.27.final_layer_norm.bias",
"model.audio.layers.27.self_attn.k_proj",
"model.audio.layers.27.self_attn.out_proj",
"model.audio.layers.27.self_attn.out_proj.bias",
"model.audio.layers.27.self_attn.q_proj",
"model.audio.layers.27.self_attn.q_proj.bias",
"model.audio.layers.27.self_attn.v_proj",
"model.audio.layers.27.self_attn.v_proj.bias",
"model.audio.layers.27.self_attn_layer_norm",
"model.audio.layers.27.self_attn_layer_norm.bias",
"model.audio.layers.28.fc1",
"model.audio.layers.28.fc1.bias",
"model.audio.layers.28.fc2",
"model.audio.layers.28.fc2.bias",
"model.audio.layers.28.final_layer_norm",
"model.audio.layers.28.final_layer_norm.bias",
"model.audio.layers.28.self_attn.k_proj",
"model.audio.layers.28.self_attn.out_proj",
"model.audio.layers.28.self_attn.out_proj.bias",
"model.audio.layers.28.self_attn.q_proj",
"model.audio.layers.28.self_attn.q_proj.bias",
"model.audio.layers.28.self_attn.v_proj",
"model.audio.layers.28.self_attn.v_proj.bias",
"model.audio.layers.28.self_attn_layer_norm",
"model.audio.layers.28.self_attn_layer_norm.bias",
"model.audio.layers.29.fc1",
"model.audio.layers.29.fc1.bias",
"model.audio.layers.29.fc2",
"model.audio.layers.29.fc2.bias",
"model.audio.layers.29.final_layer_norm",
"model.audio.layers.29.final_layer_norm.bias",
"model.audio.layers.29.self_attn.k_proj",
"model.audio.layers.29.self_attn.out_proj",
"model.audio.layers.29.self_attn.out_proj.bias",
"model.audio.layers.29.self_attn.q_proj",
"model.audio.layers.29.self_attn.q_proj.bias",
"model.audio.layers.29.self_attn.v_proj",
"model.audio.layers.29.self_attn.v_proj.bias",
"model.audio.layers.29.self_attn_layer_norm",
"model.audio.layers.29.self_attn_layer_norm.bias",
"model.audio.layers.3.fc1",
"model.audio.layers.3.fc1.bias",
"model.audio.layers.3.fc2",
"model.audio.layers.3.fc2.bias",
"model.audio.layers.3.final_layer_norm",
"model.audio.layers.3.final_layer_norm.bias",
"model.audio.layers.3.self_attn.k_proj",
"model.audio.layers.3.self_attn.out_proj",
"model.audio.layers.3.self_attn.out_proj.bias",
"model.audio.layers.3.self_attn.q_proj",
"model.audio.layers.3.self_attn.q_proj.bias",
"model.audio.layers.3.self_attn.v_proj",
"model.audio.layers.3.self_attn.v_proj.bias",
"model.audio.layers.3.self_attn_layer_norm",
"model.audio.layers.3.self_attn_layer_norm.bias",
"model.audio.layers.30.fc1",
"model.audio.layers.30.fc1.bias",
"model.audio.layers.30.fc2",
"model.audio.layers.30.fc2.bias",
"model.audio.layers.30.final_layer_norm",
"model.audio.layers.30.final_layer_norm.bias",
"model.audio.layers.30.self_attn.k_proj",
"model.audio.layers.30.self_attn.out_proj",
"model.audio.layers.30.self_attn.out_proj.bias",
"model.audio.layers.30.self_attn.q_proj",
"model.audio.layers.30.self_attn.q_proj.bias",
"model.audio.layers.30.self_attn.v_proj",
"model.audio.layers.30.self_attn.v_proj.bias",
"model.audio.layers.30.self_attn_layer_norm",
"model.audio.layers.30.self_attn_layer_norm.bias",
"model.audio.layers.31.fc1",
"model.audio.layers.31.fc1.bias",
"model.audio.layers.31.fc2",
"model.audio.layers.31.fc2.bias",
"model.audio.layers.31.final_layer_norm",
"model.audio.layers.31.final_layer_norm.bias",
"model.audio.layers.31.self_attn.k_proj",
"model.audio.layers.31.self_attn.out_proj",
"model.audio.layers.31.self_attn.out_proj.bias",
"model.audio.layers.31.self_attn.q_proj",
"model.audio.layers.31.self_attn.q_proj.bias",
"model.audio.layers.31.self_attn.v_proj",
"model.audio.layers.31.self_attn.v_proj.bias",
"model.audio.layers.31.self_attn_layer_norm",
"model.audio.layers.31.self_attn_layer_norm.bias",
"model.audio.layers.4.fc1",
"model.audio.layers.4.fc1.bias",
"model.audio.layers.4.fc2",
"model.audio.layers.4.fc2.bias",
"model.audio.layers.4.final_layer_norm",
"model.audio.layers.4.final_layer_norm.bias",
"model.audio.layers.4.self_attn.k_proj",
"model.audio.layers.4.self_attn.out_proj",
"model.audio.layers.4.self_attn.out_proj.bias",
"model.audio.layers.4.self_attn.q_proj",
"model.audio.layers.4.self_attn.q_proj.bias",
"model.audio.layers.4.self_attn.v_proj",
"model.audio.layers.4.self_attn.v_proj.bias",
"model.audio.layers.4.self_attn_layer_norm",
"model.audio.layers.4.self_attn_layer_norm.bias",
"model.audio.layers.5.fc1",
"model.audio.layers.5.fc1.bias",
"model.audio.layers.5.fc2",
"model.audio.layers.5.fc2.bias",
"model.audio.layers.5.final_layer_norm",
"model.audio.layers.5.final_layer_norm.bias",
"model.audio.layers.5.self_attn.k_proj",
"model.audio.layers.5.self_attn.out_proj",
"model.audio.layers.5.self_attn.out_proj.bias",
"model.audio.layers.5.self_attn.q_proj",
"model.audio.layers.5.self_attn.q_proj.bias",
"model.audio.layers.5.self_attn.v_proj",
"model.audio.layers.5.self_attn.v_proj.bias",
"model.audio.layers.5.self_attn_layer_norm",
"model.audio.layers.5.self_attn_layer_norm.bias",
"model.audio.layers.6.fc1",
"model.audio.layers.6.fc1.bias",
"model.audio.layers.6.fc2",
"model.audio.layers.6.fc2.bias",
"model.audio.layers.6.final_layer_norm",
"model.audio.layers.6.final_layer_norm.bias",
"model.audio.layers.6.self_attn.k_proj",
"model.audio.layers.6.self_attn.out_proj",
"model.audio.layers.6.self_attn.out_proj.bias",
"model.audio.layers.6.self_attn.q_proj",
"model.audio.layers.6.self_attn.q_proj.bias",
"model.audio.layers.6.self_attn.v_proj",
"model.audio.layers.6.self_attn.v_proj.bias",
"model.audio.layers.6.self_attn_layer_norm",
"model.audio.layers.6.self_attn_layer_norm.bias",
"model.audio.layers.7.fc1",
"model.audio.layers.7.fc1.bias",
"model.audio.layers.7.fc2",
"model.audio.layers.7.fc2.bias",
"model.audio.layers.7.final_layer_norm",
"model.audio.layers.7.final_layer_norm.bias",
"model.audio.layers.7.self_attn.k_proj",
"model.audio.layers.7.self_attn.out_proj",
"model.audio.layers.7.self_attn.out_proj.bias",
"model.audio.layers.7.self_attn.q_proj",
"model.audio.layers.7.self_attn.q_proj.bias",
"model.audio.layers.7.self_attn.v_proj",
"model.audio.layers.7.self_attn.v_proj.bias",
"model.audio.layers.7.self_attn_layer_norm",
"model.audio.layers.7.self_attn_layer_norm.bias",
"model.audio.layers.8.fc1",
"model.audio.layers.8.fc1.bias",
"model.audio.layers.8.fc2",
"model.audio.layers.8.fc2.bias",
"model.audio.layers.8.final_layer_norm",
"model.audio.layers.8.final_layer_norm.bias",
"model.audio.layers.8.self_attn.k_proj",
"model.audio.layers.8.self_attn.out_proj",
"model.audio.layers.8.self_attn.out_proj.bias",
"model.audio.layers.8.self_attn.q_proj",
"model.audio.layers.8.self_attn.q_proj.bias",
"model.audio.layers.8.self_attn.v_proj",
"model.audio.layers.8.self_attn.v_proj.bias",
"model.audio.layers.8.self_attn_layer_norm",
"model.audio.layers.8.self_attn_layer_norm.bias",
"model.audio.layers.9.fc1",
"model.audio.layers.9.fc1.bias",
"model.audio.layers.9.fc2",
"model.audio.layers.9.fc2.bias",
"model.audio.layers.9.final_layer_norm",
"model.audio.layers.9.final_layer_norm.bias",
"model.audio.layers.9.self_attn.k_proj",
"model.audio.layers.9.self_attn.out_proj",
"model.audio.layers.9.self_attn.out_proj.bias",
"model.audio.layers.9.self_attn.q_proj",
"model.audio.layers.9.self_attn.q_proj.bias",
"model.audio.layers.9.self_attn.v_proj",
"model.audio.layers.9.self_attn.v_proj.bias",
"model.audio.layers.9.self_attn_layer_norm",
"model.audio.layers.9.self_attn_layer_norm.bias",
"model.audio.q_tokens",
"model.audio.qformer.bert.embeddings.LayerNorm.beta",
"model.audio.qformer.bert.embeddings.LayerNorm.gamma",
"model.audio.qformer.bert.embeddings.position_ids",
"model.audio.qformer.bert.encoder.layer.0.attention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.attention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.attention.output.dense",
"model.audio.qformer.bert.encoder.layer.0.attention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.key",
"model.audio.qformer.bert.encoder.layer.0.attention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.query",
"model.audio.qformer.bert.encoder.layer.0.attention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.value",
"model.audio.qformer.bert.encoder.layer.0.attention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.dense",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.key",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.query",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.value",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.0.intermediate_query.dense",
"model.audio.qformer.bert.encoder.layer.0.intermediate_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.output_query.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.output_query.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.output_query.dense",
"model.audio.qformer.bert.encoder.layer.0.output_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.attention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.attention.output.dense",
"model.audio.qformer.bert.encoder.layer.1.attention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.key",
"model.audio.qformer.bert.encoder.layer.1.attention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.query",
"model.audio.qformer.bert.encoder.layer.1.attention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.value",
"model.audio.qformer.bert.encoder.layer.1.attention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.dense",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.key",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.query",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.value",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.1.intermediate_query.dense",
"model.audio.qformer.bert.encoder.layer.1.intermediate_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.output_query.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.output_query.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.output_query.dense",
"model.audio.qformer.bert.encoder.layer.1.output_query.dense.bias",
"model.language_model.embed_tokens",
"model.language_model.layers.0.linear_attn.conv1d",
"model.language_model.layers.0.linear_attn.in_proj_a",
"model.language_model.layers.0.linear_attn.in_proj_b",
"model.language_model.layers.1.linear_attn.conv1d",
"model.language_model.layers.1.linear_attn.in_proj_a",
"model.language_model.layers.1.linear_attn.in_proj_b",
"model.language_model.layers.10.linear_attn.conv1d",
"model.language_model.layers.10.linear_attn.in_proj_a",
"model.language_model.layers.10.linear_attn.in_proj_b",
"model.language_model.layers.12.linear_attn.conv1d",
"model.language_model.layers.12.linear_attn.in_proj_a",
"model.language_model.layers.12.linear_attn.in_proj_b",
"model.language_model.layers.13.linear_attn.conv1d",
"model.language_model.layers.13.linear_attn.in_proj_a",
"model.language_model.layers.13.linear_attn.in_proj_b",
"model.language_model.layers.14.linear_attn.conv1d",
"model.language_model.layers.14.linear_attn.in_proj_a",
"model.language_model.layers.14.linear_attn.in_proj_b",
"model.language_model.layers.16.linear_attn.conv1d",
"model.language_model.layers.16.linear_attn.in_proj_a",
"model.language_model.layers.16.linear_attn.in_proj_b",
"model.language_model.layers.17.linear_attn.conv1d",
"model.language_model.layers.17.linear_attn.in_proj_a",
"model.language_model.layers.17.linear_attn.in_proj_b",
"model.language_model.layers.18.linear_attn.conv1d",
"model.language_model.layers.18.linear_attn.in_proj_a",
"model.language_model.layers.18.linear_attn.in_proj_b",
"model.language_model.layers.2.linear_attn.conv1d",
"model.language_model.layers.2.linear_attn.in_proj_a",
"model.language_model.layers.2.linear_attn.in_proj_b",
"model.language_model.layers.20.linear_attn.conv1d",
"model.language_model.layers.20.linear_attn.in_proj_a",
"model.language_model.layers.20.linear_attn.in_proj_b",
"model.language_model.layers.21.linear_attn.conv1d",
"model.language_model.layers.21.linear_attn.in_proj_a",
"model.language_model.layers.21.linear_attn.in_proj_b",
"model.language_model.layers.22.linear_attn.conv1d",
"model.language_model.layers.22.linear_attn.in_proj_a",
"model.language_model.layers.22.linear_attn.in_proj_b",
"model.language_model.layers.24.linear_attn.conv1d",
"model.language_model.layers.24.linear_attn.in_proj_a",
"model.language_model.layers.24.linear_attn.in_proj_b",
"model.language_model.layers.25.linear_attn.conv1d",
"model.language_model.layers.25.linear_attn.in_proj_a",
"model.language_model.layers.25.linear_attn.in_proj_b",
"model.language_model.layers.26.linear_attn.conv1d",
"model.language_model.layers.26.linear_attn.in_proj_a",
"model.language_model.layers.26.linear_attn.in_proj_b",
"model.language_model.layers.28.linear_attn.conv1d",
"model.language_model.layers.28.linear_attn.in_proj_a",
"model.language_model.layers.28.linear_attn.in_proj_b",
"model.language_model.layers.29.linear_attn.conv1d",
"model.language_model.layers.29.linear_attn.in_proj_a",
"model.language_model.layers.29.linear_attn.in_proj_b",
"model.language_model.layers.30.linear_attn.conv1d",
"model.language_model.layers.30.linear_attn.in_proj_a",
"model.language_model.layers.30.linear_attn.in_proj_b",
"model.language_model.layers.4.linear_attn.conv1d",
"model.language_model.layers.4.linear_attn.in_proj_a",
"model.language_model.layers.4.linear_attn.in_proj_b",
"model.language_model.layers.5.linear_attn.conv1d",
"model.language_model.layers.5.linear_attn.in_proj_a",
"model.language_model.layers.5.linear_attn.in_proj_b",
"model.language_model.layers.6.linear_attn.conv1d",
"model.language_model.layers.6.linear_attn.in_proj_a",
"model.language_model.layers.6.linear_attn.in_proj_b",
"model.language_model.layers.8.linear_attn.conv1d",
"model.language_model.layers.8.linear_attn.in_proj_a",
"model.language_model.layers.8.linear_attn.in_proj_b",
"model.language_model.layers.9.linear_attn.conv1d",
"model.language_model.layers.9.linear_attn.in_proj_a",
"model.language_model.layers.9.linear_attn.in_proj_b",
"model.language_model.norm",
"model.visual",
"model.visual.blocks.0.attn.proj",
"model.visual.blocks.0.attn.proj.bias",
"model.visual.blocks.0.attn.qkv",
"model.visual.blocks.0.attn.qkv.bias",
"model.visual.blocks.0.mlp.linear_fc1",
"model.visual.blocks.0.mlp.linear_fc1.bias",
"model.visual.blocks.0.mlp.linear_fc2",
"model.visual.blocks.0.mlp.linear_fc2.bias",
"model.visual.blocks.0.norm1",
"model.visual.blocks.0.norm1.bias",
"model.visual.blocks.0.norm2",
"model.visual.blocks.0.norm2.bias",
"model.visual.blocks.1.attn.proj",
"model.visual.blocks.1.attn.proj.bias",
"model.visual.blocks.1.attn.qkv",
"model.visual.blocks.1.attn.qkv.bias",
"model.visual.blocks.1.mlp.linear_fc1",
"model.visual.blocks.1.mlp.linear_fc1.bias",
"model.visual.blocks.1.mlp.linear_fc2",
"model.visual.blocks.1.mlp.linear_fc2.bias",
"model.visual.blocks.1.norm1",
"model.visual.blocks.1.norm1.bias",
"model.visual.blocks.1.norm2",
"model.visual.blocks.1.norm2.bias",
"model.visual.blocks.10.attn.proj",
"model.visual.blocks.10.attn.proj.bias",
"model.visual.blocks.10.attn.qkv",
"model.visual.blocks.10.attn.qkv.bias",
"model.visual.blocks.10.mlp.linear_fc1",
"model.visual.blocks.10.mlp.linear_fc1.bias",
"model.visual.blocks.10.mlp.linear_fc2",
"model.visual.blocks.10.mlp.linear_fc2.bias",
"model.visual.blocks.10.norm1",
"model.visual.blocks.10.norm1.bias",
"model.visual.blocks.10.norm2",
"model.visual.blocks.10.norm2.bias",
"model.visual.blocks.11.attn.proj",
"model.visual.blocks.11.attn.proj.bias",
"model.visual.blocks.11.attn.qkv",
"model.visual.blocks.11.attn.qkv.bias",
"model.visual.blocks.11.mlp.linear_fc1",
"model.visual.blocks.11.mlp.linear_fc1.bias",
"model.visual.blocks.11.mlp.linear_fc2",
"model.visual.blocks.11.mlp.linear_fc2.bias",
"model.visual.blocks.11.norm1",
"model.visual.blocks.11.norm1.bias",
"model.visual.blocks.11.norm2",
"model.visual.blocks.11.norm2.bias",
"model.visual.blocks.12.attn.proj",
"model.visual.blocks.12.attn.proj.bias",
"model.visual.blocks.12.attn.qkv",
"model.visual.blocks.12.attn.qkv.bias",
"model.visual.blocks.12.mlp.linear_fc1",
"model.visual.blocks.12.mlp.linear_fc1.bias",
"model.visual.blocks.12.mlp.linear_fc2",
"model.visual.blocks.12.mlp.linear_fc2.bias",
"model.visual.blocks.12.norm1",
"model.visual.blocks.12.norm1.bias",
"model.visual.blocks.12.norm2",
"model.visual.blocks.12.norm2.bias",
"model.visual.blocks.13.attn.proj",
"model.visual.blocks.13.attn.proj.bias",
"model.visual.blocks.13.attn.qkv",
"model.visual.blocks.13.attn.qkv.bias",
"model.visual.blocks.13.mlp.linear_fc1",
"model.visual.blocks.13.mlp.linear_fc1.bias",
"model.visual.blocks.13.mlp.linear_fc2",
"model.visual.blocks.13.mlp.linear_fc2.bias",
"model.visual.blocks.13.norm1",
"model.visual.blocks.13.norm1.bias",
"model.visual.blocks.13.norm2",
"model.visual.blocks.13.norm2.bias",
"model.visual.blocks.14.attn.proj",
"model.visual.blocks.14.attn.proj.bias",
"model.visual.blocks.14.attn.qkv",
"model.visual.blocks.14.attn.qkv.bias",
"model.visual.blocks.14.mlp.linear_fc1",
"model.visual.blocks.14.mlp.linear_fc1.bias",
"model.visual.blocks.14.mlp.linear_fc2",
"model.visual.blocks.14.mlp.linear_fc2.bias",
"model.visual.blocks.14.norm1",
"model.visual.blocks.14.norm1.bias",
"model.visual.blocks.14.norm2",
"model.visual.blocks.14.norm2.bias",
"model.visual.blocks.15.attn.proj",
"model.visual.blocks.15.attn.proj.bias",
"model.visual.blocks.15.attn.qkv",
"model.visual.blocks.15.attn.qkv.bias",
"model.visual.blocks.15.mlp.linear_fc1",
"model.visual.blocks.15.mlp.linear_fc1.bias",
"model.visual.blocks.15.mlp.linear_fc2",
"model.visual.blocks.15.mlp.linear_fc2.bias",
"model.visual.blocks.15.norm1",
"model.visual.blocks.15.norm1.bias",
"model.visual.blocks.15.norm2",
"model.visual.blocks.15.norm2.bias",
"model.visual.blocks.16.attn.proj",
"model.visual.blocks.16.attn.proj.bias",
"model.visual.blocks.16.attn.qkv",
"model.visual.blocks.16.attn.qkv.bias",
"model.visual.blocks.16.mlp.linear_fc1",
"model.visual.blocks.16.mlp.linear_fc1.bias",
"model.visual.blocks.16.mlp.linear_fc2",
"model.visual.blocks.16.mlp.linear_fc2.bias",
"model.visual.blocks.16.norm1",
"model.visual.blocks.16.norm1.bias",
"model.visual.blocks.16.norm2",
"model.visual.blocks.16.norm2.bias",
"model.visual.blocks.17.attn.proj",
"model.visual.blocks.17.attn.proj.bias",
"model.visual.blocks.17.attn.qkv",
"model.visual.blocks.17.attn.qkv.bias",
"model.visual.blocks.17.mlp.linear_fc1",
"model.visual.blocks.17.mlp.linear_fc1.bias",
"model.visual.blocks.17.mlp.linear_fc2",
"model.visual.blocks.17.mlp.linear_fc2.bias",
"model.visual.blocks.17.norm1",
"model.visual.blocks.17.norm1.bias",
"model.visual.blocks.17.norm2",
"model.visual.blocks.17.norm2.bias",
"model.visual.blocks.18.attn.proj",
"model.visual.blocks.18.attn.proj.bias",
"model.visual.blocks.18.attn.qkv",
"model.visual.blocks.18.attn.qkv.bias",
"model.visual.blocks.18.mlp.linear_fc1",
"model.visual.blocks.18.mlp.linear_fc1.bias",
"model.visual.blocks.18.mlp.linear_fc2",
"model.visual.blocks.18.mlp.linear_fc2.bias",
"model.visual.blocks.18.norm1",
"model.visual.blocks.18.norm1.bias",
"model.visual.blocks.18.norm2",
"model.visual.blocks.18.norm2.bias",
"model.visual.blocks.19.attn.proj",
"model.visual.blocks.19.attn.proj.bias",
"model.visual.blocks.19.attn.qkv",
"model.visual.blocks.19.attn.qkv.bias",
"model.visual.blocks.19.mlp.linear_fc1",
"model.visual.blocks.19.mlp.linear_fc1.bias",
"model.visual.blocks.19.mlp.linear_fc2",
"model.visual.blocks.19.mlp.linear_fc2.bias",
"model.visual.blocks.19.norm1",
"model.visual.blocks.19.norm1.bias",
"model.visual.blocks.19.norm2",
"model.visual.blocks.19.norm2.bias",
"model.visual.blocks.2.attn.proj",
"model.visual.blocks.2.attn.proj.bias",
"model.visual.blocks.2.attn.qkv",
"model.visual.blocks.2.attn.qkv.bias",
"model.visual.blocks.2.mlp.linear_fc1",
"model.visual.blocks.2.mlp.linear_fc1.bias",
"model.visual.blocks.2.mlp.linear_fc2",
"model.visual.blocks.2.mlp.linear_fc2.bias",
"model.visual.blocks.2.norm1",
"model.visual.blocks.2.norm1.bias",
"model.visual.blocks.2.norm2",
"model.visual.blocks.2.norm2.bias",
"model.visual.blocks.20.attn.proj",
"model.visual.blocks.20.attn.proj.bias",
"model.visual.blocks.20.attn.qkv",
"model.visual.blocks.20.attn.qkv.bias",
"model.visual.blocks.20.mlp.linear_fc1",
"model.visual.blocks.20.mlp.linear_fc1.bias",
"model.visual.blocks.20.mlp.linear_fc2",
"model.visual.blocks.20.mlp.linear_fc2.bias",
"model.visual.blocks.20.norm1",
"model.visual.blocks.20.norm1.bias",
"model.visual.blocks.20.norm2",
"model.visual.blocks.20.norm2.bias",
"model.visual.blocks.21.attn.proj",
"model.visual.blocks.21.attn.proj.bias",
"model.visual.blocks.21.attn.qkv",
"model.visual.blocks.21.attn.qkv.bias",
"model.visual.blocks.21.mlp.linear_fc1",
"model.visual.blocks.21.mlp.linear_fc1.bias",
"model.visual.blocks.21.mlp.linear_fc2",
"model.visual.blocks.21.mlp.linear_fc2.bias",
"model.visual.blocks.21.norm1",
"model.visual.blocks.21.norm1.bias",
"model.visual.blocks.21.norm2",
"model.visual.blocks.21.norm2.bias",
"model.visual.blocks.22.attn.proj",
"model.visual.blocks.22.attn.proj.bias",
"model.visual.blocks.22.attn.qkv",
"model.visual.blocks.22.attn.qkv.bias",
"model.visual.blocks.22.mlp.linear_fc1",
"model.visual.blocks.22.mlp.linear_fc1.bias",
"model.visual.blocks.22.mlp.linear_fc2",
"model.visual.blocks.22.mlp.linear_fc2.bias",
"model.visual.blocks.22.norm1",
"model.visual.blocks.22.norm1.bias",
"model.visual.blocks.22.norm2",
"model.visual.blocks.22.norm2.bias",
"model.visual.blocks.23.attn.proj",
"model.visual.blocks.23.attn.proj.bias",
"model.visual.blocks.23.attn.qkv",
"model.visual.blocks.23.attn.qkv.bias",
"model.visual.blocks.23.mlp.linear_fc1",
"model.visual.blocks.23.mlp.linear_fc1.bias",
"model.visual.blocks.23.mlp.linear_fc2",
"model.visual.blocks.23.mlp.linear_fc2.bias",
"model.visual.blocks.23.norm1",
"model.visual.blocks.23.norm1.bias",
"model.visual.blocks.23.norm2",
"model.visual.blocks.23.norm2.bias",
"model.visual.blocks.24.attn.proj",
"model.visual.blocks.24.attn.proj.bias",
"model.visual.blocks.24.attn.qkv",
"model.visual.blocks.24.attn.qkv.bias",
"model.visual.blocks.24.mlp.linear_fc1",
"model.visual.blocks.24.mlp.linear_fc1.bias",
"model.visual.blocks.24.mlp.linear_fc2",
"model.visual.blocks.24.mlp.linear_fc2.bias",
"model.visual.blocks.24.norm1",
"model.visual.blocks.24.norm1.bias",
"model.visual.blocks.24.norm2",
"model.visual.blocks.24.norm2.bias",
"model.visual.blocks.25.attn.proj",
"model.visual.blocks.25.attn.proj.bias",
"model.visual.blocks.25.attn.qkv",
"model.visual.blocks.25.attn.qkv.bias",
"model.visual.blocks.25.mlp.linear_fc1",
"model.visual.blocks.25.mlp.linear_fc1.bias",
"model.visual.blocks.25.mlp.linear_fc2",
"model.visual.blocks.25.mlp.linear_fc2.bias",
"model.visual.blocks.25.norm1",
"model.visual.blocks.25.norm1.bias",
"model.visual.blocks.25.norm2",
"model.visual.blocks.25.norm2.bias",
"model.visual.blocks.26.attn.proj",
"model.visual.blocks.26.attn.proj.bias",
"model.visual.blocks.26.attn.qkv",
"model.visual.blocks.26.attn.qkv.bias",
"model.visual.blocks.26.mlp.linear_fc1",
"model.visual.blocks.26.mlp.linear_fc1.bias",
"model.visual.blocks.26.mlp.linear_fc2",
"model.visual.blocks.26.mlp.linear_fc2.bias",
"model.visual.blocks.26.norm1",
"model.visual.blocks.26.norm1.bias",
"model.visual.blocks.26.norm2",
"model.visual.blocks.26.norm2.bias",
"model.visual.blocks.3.attn.proj",
"model.visual.blocks.3.attn.proj.bias",
"model.visual.blocks.3.attn.qkv",
"model.visual.blocks.3.attn.qkv.bias",
"model.visual.blocks.3.mlp.linear_fc1",
"model.visual.blocks.3.mlp.linear_fc1.bias",
"model.visual.blocks.3.mlp.linear_fc2",
"model.visual.blocks.3.mlp.linear_fc2.bias",
"model.visual.blocks.3.norm1",
"model.visual.blocks.3.norm1.bias",
"model.visual.blocks.3.norm2",
"model.visual.blocks.3.norm2.bias",
"model.visual.blocks.4.attn.proj",
"model.visual.blocks.4.attn.proj.bias",
"model.visual.blocks.4.attn.qkv",
"model.visual.blocks.4.attn.qkv.bias",
"model.visual.blocks.4.mlp.linear_fc1",
"model.visual.blocks.4.mlp.linear_fc1.bias",
"model.visual.blocks.4.mlp.linear_fc2",
"model.visual.blocks.4.mlp.linear_fc2.bias",
"model.visual.blocks.4.norm1",
"model.visual.blocks.4.norm1.bias",
"model.visual.blocks.4.norm2",
"model.visual.blocks.4.norm2.bias",
"model.visual.blocks.5.attn.proj",
"model.visual.blocks.5.attn.proj.bias",
"model.visual.blocks.5.attn.qkv",
"model.visual.blocks.5.attn.qkv.bias",
"model.visual.blocks.5.mlp.linear_fc1",
"model.visual.blocks.5.mlp.linear_fc1.bias",
"model.visual.blocks.5.mlp.linear_fc2",
"model.visual.blocks.5.mlp.linear_fc2.bias",
"model.visual.blocks.5.norm1",
"model.visual.blocks.5.norm1.bias",
"model.visual.blocks.5.norm2",
"model.visual.blocks.5.norm2.bias",
"model.visual.blocks.6.attn.proj",
"model.visual.blocks.6.attn.proj.bias",
"model.visual.blocks.6.attn.qkv",
"model.visual.blocks.6.attn.qkv.bias",
"model.visual.blocks.6.mlp.linear_fc1",
"model.visual.blocks.6.mlp.linear_fc1.bias",
"model.visual.blocks.6.mlp.linear_fc2",
"model.visual.blocks.6.mlp.linear_fc2.bias",
"model.visual.blocks.6.norm1",
"model.visual.blocks.6.norm1.bias",
"model.visual.blocks.6.norm2",
"model.visual.blocks.6.norm2.bias",
"model.visual.blocks.7.attn.proj",
"model.visual.blocks.7.attn.proj.bias",
"model.visual.blocks.7.attn.qkv",
"model.visual.blocks.7.attn.qkv.bias",
"model.visual.blocks.7.mlp.linear_fc1",
"model.visual.blocks.7.mlp.linear_fc1.bias",
"model.visual.blocks.7.mlp.linear_fc2",
"model.visual.blocks.7.mlp.linear_fc2.bias",
"model.visual.blocks.7.norm1",
"model.visual.blocks.7.norm1.bias",
"model.visual.blocks.7.norm2",
"model.visual.blocks.7.norm2.bias",
"model.visual.blocks.8.attn.proj",
"model.visual.blocks.8.attn.proj.bias",
"model.visual.blocks.8.attn.qkv",
"model.visual.blocks.8.attn.qkv.bias",
"model.visual.blocks.8.mlp.linear_fc1",
"model.visual.blocks.8.mlp.linear_fc1.bias",
"model.visual.blocks.8.mlp.linear_fc2",
"model.visual.blocks.8.mlp.linear_fc2.bias",
"model.visual.blocks.8.norm1",
"model.visual.blocks.8.norm1.bias",
"model.visual.blocks.8.norm2",
"model.visual.blocks.8.norm2.bias",
"model.visual.blocks.9.attn.proj",
"model.visual.blocks.9.attn.proj.bias",
"model.visual.blocks.9.attn.qkv",
"model.visual.blocks.9.attn.qkv.bias",
"model.visual.blocks.9.mlp.linear_fc1",
"model.visual.blocks.9.mlp.linear_fc1.bias",
"model.visual.blocks.9.mlp.linear_fc2",
"model.visual.blocks.9.mlp.linear_fc2.bias",
"model.visual.blocks.9.norm1",
"model.visual.blocks.9.norm1.bias",
"model.visual.blocks.9.norm2",
"model.visual.blocks.9.norm2.bias",
"model.visual.merger.linear_fc1",
"model.visual.merger.linear_fc1.bias",
"model.visual.merger.linear_fc2",
"model.visual.merger.linear_fc2.bias",
"model.visual.merger.norm",
"model.visual.merger.norm.bias",
"model.visual.patch_embed.proj",
"model.visual.patch_embed.proj.bias",
"model.visual.pos_embed",
"mtp"
]
}
}
{
"architectures": [
"WeMMVideoForConditionalGeneration"
],
"audio_config": {
"activation_dropout": 0.0,
"activation_function": "gelu",
"apply_spec_augment": false,
"attention_dropout": 0.0,
"bos_token_id": 50257,
"classifier_proj_size": 256,
"d_model": 1280,
"decoder_attention_heads": 20,
"decoder_ffn_dim": 5120,
"decoder_layerdrop": 0.0,
"decoder_layers": 32,
"decoder_start_token_id": 50258,
"dropout": 0.0,
"dtype": "bfloat16",
"encoder_attention_heads": 20,
"encoder_ffn_dim": 5120,
"encoder_layerdrop": 0.0,
"encoder_layers": 32,
"eos_token_id": 50257,
"init_std": 0.02,
"mask_feature_length": 10,
"mask_feature_min_masks": 0,
"mask_feature_prob": 0.0,
"mask_time_length": 10,
"mask_time_min_masks": 2,
"mask_time_prob": 0.05,
"max_source_positions": 1500,
"max_target_positions": 448,
"median_filter_width": 7,
"model_type": "whisper",
"num_hidden_layers": 32,
"num_mel_bins": 128,
"pad_token_id": 50256,
"qformer_config": {
"add_cross_attention": true,
"architectures": null,
"attention_probs_dropout_prob": 0.1,
"chunk_size_feed_forward": 0,
"classifier_dropout": null,
"cross_attention_freq": 1,
"dtype": null,
"encoder_width": 1280,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 4104,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"initializer_range": 0.02,
"intermediate_size": 12288,
"is_encoder_decoder": false,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"layer_norm_eps": 1e-12,
"max_position_embeddings": 512,
"model_type": "bert",
"num_attention_heads": 12,
"num_hidden_layers": 2,
"out_hidden_size": 4096,
"output_attentions": false,
"output_hidden_states": false,
"pad_token_id": 0,
"position_embedding_type": "absolute",
"problem_type": null,
"query_length": 1,
"return_dict": true,
"type_vocab_size": 2,
"use_cache": true,
"vocab_size": 30522
},
"scale_embedding": false,
"second_per_window": 0.1,
"second_stride": 0.1,
"tie_word_embeddings": true,
"use_cache": true,
"use_weighted_layer_sum": false,
"vocab_size": 51866
},
"audio_token_id": 248076,
"dtype": "bfloat16",
"image_token_id": 248056,
"model_type": "qwen3_5",
"text_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"attn_output_gate": true,
"bos_token_id": null,
"dtype": "bfloat16",
"eos_token_id": 248044,
"full_attention_interval": 4,
"head_dim": 256,
"hidden_act": "silu",