config.json

{
  "_name_or_path": "",
  "architectures": [
    "OmniSpeechVisionQwenForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "freeze_mm_mlp_adapter": false,
  "freeze_mm_vision_resampler": false,
  "freeze_mm_vision_tower": true,
  "freeze_speech_adapter": false,
  "hidden_act": "silu",
  "hidden_size": 3584,
  "initializer_range": 0.02,
  "intermediate_size": 18944,
  "max_position_embeddings": 32768,
  "max_window_layers": 28,
  "mm_hidden_size": 1152,
  "mm_projector_type": "ola_mlp",
  "mm_use_im_patch_token": false,
  "mm_use_im_start_end": false,
  "mm_vision_select_feature": "patch",
  "mm_vision_select_layer": -1,
  "music_encoder": "./pretrained/BEATs_iter3_plus_AS2M_finetuned_on_AS2M_cpt2.pt",
  "mm_vision_tower": "./pretrained/oryx_vit.pth",
  "mm_vision_tower_lr": null,
  "modality_max_length": "None",
  "model_type": "ola_qwen",
  "num_attention_heads": 28,
  "num_hidden_layers": 28,
  "num_key_value_heads": 4,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "speech_encoder": "./pretrained/large-v3.pt",
  "speech_encoder_ds_rate": 10,
  "speech_encoder_hidden_size": 2048,
  "speech_encoder_type": "dual",
  "speech_projector_type": "linear",
  "tie_word_embeddings": false,
  "tokenizer_model_max_length": 16384,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.43.4",
  "tune_mm_mlp_adapter": false,
  "tune_mm_vision_resampler": false,
  "tune_speech_adapter": false,
  "unfreeze_mm_vision_tower": false,
  "use_cache": true,
  "use_mm_proj": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}