config.json
| 1 | { |
| 2 | "model_type": "moss_tts_delay", |
| 3 | "architectures": [ |
| 4 | "MossTTSDelayModel" |
| 5 | ], |
| 6 | "auto_map": { |
| 7 | "AutoConfig": "configuration_moss_tts.MossTTSDelayConfig", |
| 8 | "AutoModel": "modeling_moss_tts.MossTTSDelayModel" |
| 9 | }, |
| 10 | "dtype": "bfloat16", |
| 11 | "initializer_range": 0.02, |
| 12 | "language_config": { |
| 13 | "_name_or_path": "Qwen/Qwen3-8B", |
| 14 | "architectures": [ |
| 15 | "Qwen3ForCausalLM" |
| 16 | ], |
| 17 | "attention_bias": false, |
| 18 | "attention_dropout": 0.0, |
| 19 | "bos_token_id": 151643, |
| 20 | "eos_token_id": 151645, |
| 21 | "pad_token_id": 151643, |
| 22 | "head_dim": 128, |
| 23 | "hidden_act": "silu", |
| 24 | "hidden_size": 4096, |
| 25 | "initializer_range": 0.02, |
| 26 | "intermediate_size": 12288, |
| 27 | "layer_types": [ |
| 28 | "full_attention", |
| 29 | "full_attention", |
| 30 | "full_attention", |
| 31 | "full_attention", |
| 32 | "full_attention", |
| 33 | "full_attention", |
| 34 | "full_attention", |
| 35 | "full_attention", |
| 36 | "full_attention", |
| 37 | "full_attention", |
| 38 | "full_attention", |
| 39 | "full_attention", |
| 40 | "full_attention", |
| 41 | "full_attention", |
| 42 | "full_attention", |
| 43 | "full_attention", |
| 44 | "full_attention", |
| 45 | "full_attention", |
| 46 | "full_attention", |
| 47 | "full_attention", |
| 48 | "full_attention", |
| 49 | "full_attention", |
| 50 | "full_attention", |
| 51 | "full_attention", |
| 52 | "full_attention", |
| 53 | "full_attention", |
| 54 | "full_attention", |
| 55 | "full_attention", |
| 56 | "full_attention", |
| 57 | "full_attention", |
| 58 | "full_attention", |
| 59 | "full_attention", |
| 60 | "full_attention", |
| 61 | "full_attention", |
| 62 | "full_attention", |
| 63 | "full_attention" |
| 64 | ], |
| 65 | "max_position_embeddings": 40960, |
| 66 | "max_window_layers": 36, |
| 67 | "model_type": "qwen3", |
| 68 | "num_attention_heads": 32, |
| 69 | "num_hidden_layers": 36, |
| 70 | "num_key_value_heads": 8, |
| 71 | "rms_norm_eps": 1e-06, |
| 72 | "rope_scaling": null, |
| 73 | "rope_theta": 1000000, |
| 74 | "sliding_window": null, |
| 75 | "use_cache": true, |
| 76 | "use_sliding_window": false, |
| 77 | "vocab_size": 155648 |
| 78 | }, |
| 79 | "n_vq": 32, |
| 80 | "audio_vocab_size": 1024, |
| 81 | "audio_user_slot_token_id": 151654, |
| 82 | "audio_assistant_gen_slot_token_id": 151656, |
| 83 | "audio_assistant_delay_slot_token_id": 151662, |
| 84 | "audio_start_token_id": 151652, |
| 85 | "audio_end_token_id": 151653, |
| 86 | "audio_pad_code": 1024, |
| 87 | "sampling_rate": 24000, |
| 88 | "transformers_version": "4.57.1" |
| 89 | } |
| 90 | |