{
  "architectures": [
    "LlamaForCasualLMWithXBridge"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "dec_lambda": 1.0,
  "eos_token_id": 2,
  "freeze_dec": true,
  "freeze_enc": true,
  "freeze_llm": true,
  "freeze_mapping_enc2llm": false,
  "freeze_mapping_llm2dec": false,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "llm_bos_token_id": 128000,
  "llm_eos_token_id": 128001,
  "llm_pad_token_id": 128002,
  "llm_path": "/data/bumengyu/2.model/llama3-8B",
  "max_gen_len": 128,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "mt_eos_token_id": 2,
  "mt_pad_token_id": 1,
  "mt_path": "/data/bumengyu/2.model/nllb-200-distilled-1.3B",
  "mt_vocab_size": 256206,
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "ot_lambda": 6.0,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.45.2",
  "use_cache": true,
  "vocab_size": 32000
}