{
  "_name_or_path": "LLaMA-MoD",
  "architectures": [
    "LlamaMoDForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_llama.LlamaMoDConfig",
    "AutoModel": "modeling_llama.MiniCPMModelLlamaMoDModel",
    "AutoModelForCausalLM": "modeling_llama.LlamaMoDForCausalLM"
  },
  "bos_token_id": 1,
  "capacity_load": 0.5,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mod_method": "mod_dual",
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "router_aux_loss_coef": 0.005,
  "setup_layer_mod": [
    false,
    false,
    false,
    false,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    true,
    false,
    false
  ],
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.40.0.dev0",
  "use_cache": true,
  "vocab_size": 32000
}