univoice_all/config.json

{
  "_name_or_path": "hf_ckpts/SmolLM2-360M-Instruct",
  "add_pos_embed_each_layer": false,
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "audio_encoder_path": "hf_ckpts/whisper-large-v3",
  "bos_token_id": 1,
  "decoder_t_embed": "add_before_speech_tokens",
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 960,
  "initializer_range": 0.02,
  "intermediate_size": 2560,
  "is_llama_config": true,
  "learn_sigma": true,
  "max_position_embeddings": 8192,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 15,
  "num_hidden_layers": 32,
  "num_key_value_heads": 5,
  "pad_token_id": 2,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_interleaved": false,
  "rope_scaling": null,
  "rope_theta": 100000,
  "speaker_encoder_path": "hf_ckpts/wav2vec2-large-xlsr-53",
  "tie_word_embeddings": true,
  "tokenizer_max_length": 1024,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers.js_config": {
    "kv_cache_dtype": {
      "fp16": "float16",
      "q4f16": "float16"
    }
  },
  "transformers_version": "4.39.0",
  "use_adaln_final_layer": true,
  "use_bi_attn_speech_tokens": true,
  "use_cache": true,
  "use_flash_attn": false,
  "use_hybrid_attn_mask": false,
  "use_pos_embed": true,
  "vocab_size": 49152
}