UniMoE-Audio-Preview / config.json
foggyforest's picture
Upload 14 files
e576ca4 verified
{
"architectures": [
"UniAudioRVQQwen2_5VLMoEForConditionalGeneration"
],
"auto_map": {
"AutoConfig": "modeling.UniMoEAudioConfig",
"AutoModelForCausalLM": "modeling.UniMoEAudio"
},
"attention_dropout": 0.0,
"bos_token_id": 151643,
"codec_bos_value": 1026,
"codec_channels": 12,
"codec_delay_pattern": [
0,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18
],
"codec_eos_value": 1024,
"codec_pad_value": 1025,
"codec_placeholder_value": 151665,
"codec_vocab_size": 1027,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 2048,
"image_token_id": 151655,
"initializer_range": 0.02,
"intermediate_size": 11008,
"max_position_embeddings": 128000,
"max_window_layers": 70,
"model_type": "uni_audio_rvq_qwen2_5vl_moe",
"num_attention_heads": 16,
"num_hidden_layers": 36,
"num_key_value_heads": 2,
"rms_norm_eps": 1e-06,
"rope_scaling": {
"mrope_section": [
16,
24,
24
],
"rope_type": "default",
"type": "default"
},
"rope_theta": 1000000.0,
"sliding_window": 32768,
"text_config": {
"architectures": [
"Qwen2_5_VLForConditionalGeneration"
],
"attention_dropout": 0.0,
"avg_hidden_states_last": false,
"bos_token_id": 151643,
"capacity_factor": 6.0,
"drop_policy": "probs",
"drop_token_num_print": true,
"dynamic_intermediate_size": 2752,
"enable_expert_tensor_parallelism": false,
"eos_token_id": 151645,
"ep_size": 1,
"fixed_ep_size": 1,
"fp32_gate": true,
"hidden_act": "silu",
"hidden_size": 2048,
"ignore_differentiable_router": true,
"image_token_id": null,
"initializer_range": 0.02,
"input_jitter_noise": 0.01,
"intermediate_size": 11008,
"l_aux_weight": 0.025,
"l_aux_weight_decay_steps": 10000,
"layer_types": [
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention"
],
"max_position_embeddings": 128000,
"max_window_layers": 70,
"min_capacity": 8,
"min_l_aux_weight": 0.001,
"mlp_dynamic_expert_num": 8,
"mlp_dynamic_null_expert_num": 1,
"mlp_dynamic_top_k": 0.0,
"mlp_dynamic_top_p": 0.7,
"mlp_fixed_expert_num": 2,
"model_type": "qwen2_5_vl_moe_text",
"num_attention_heads": 16,
"num_hidden_layers": 36,
"num_key_value_heads": 2,
"rms_norm_eps": 1e-06,
"rope_scaling": {
"mrope_section": [
16,
24,
24
],
"rope_type": "default",
"type": "default"
},
"rope_theta": 1000000.0,
"router_jitter_noise": 0.01,
"shared_intermediate_size": 1376,
"sliding_window": null,
"token_drop": false,
"torch_dtype": "bfloat16",
"use_cache": true,
"use_sliding_window": false,
"video_token_id": null,
"vision_end_token_id": 151653,
"vision_start_token_id": 151652,
"vision_token_id": 151654,
"vocab_size": 151676
},
"torch_dtype": "bfloat16",
"transformers_version": "4.53.1",
"use_cache": true,
"use_sliding_window": false,
"video_token_id": 151656,
"vision_config": {
"depth": 32,
"fullatt_block_indexes": [
7,
15,
23,
31
],
"hidden_act": "silu",
"hidden_size": 1280,
"in_channels": 3,
"in_chans": 3,
"initializer_range": 0.02,
"intermediate_size": 3420,
"model_type": "qwen2_5_vl",
"num_heads": 16,
"out_hidden_size": 2048,
"patch_size": 14,
"spatial_merge_size": 2,
"spatial_patch_size": 14,
"temporal_patch_size": 2,
"tokens_per_second": 2,
"torch_dtype": "bfloat16",
"window_size": 112
},
"vision_end_token_id": 151653,
"vision_start_token_id": 151652,
"vision_token_id": 151654,
"vocab_size": 151936
}