{
  "_name_or_path": "./out",
  "architectures": [
    "DeepseekForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_deepseek.DeepseekConfig",
    "AutoModel": "modeling_deepseek.DeepseekModel",
    "AutoModelForCausalLM": "modeling_deepseek.DeepseekForCausalLM"
  },
  "aux_loss_alpha": 0.02,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "first_k_dense_replace": 0,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 14336,
  "max_position_embeddings": 32768,
  "model_type": "deepseek",
  "moe_intermediate_size": 14336,
  "moe_layer_freq": 1,
  "n_routed_experts": 8,
  "n_shared_experts": null,
  "norm_topk_prob": true,
  "num_attention_heads": 32,
  "num_experts_per_tok": 2,
  "num_hidden_layers": 32,
  "num_key_value_heads": 8,
  "num_local_experts": 8,
  "output_router_logits": false,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "router_aux_loss_coef": 0.02,
  "scoring_func": "softmax",
  "seq_aux": true,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.40.0.dev0",
  "use_cache": false,
  "vocab_size": 32000
}