{ "_name_or_path": "/app/mergekit/merged", "architectures": [ "DeepseekForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "auto_map": { "AutoConfig": "configuration_deepseek.DeepseekConfig", "AutoModel": "deepseek-ai/deepseek-moe-16b-base--modeling_deepseek.DeepseekModel", "AutoModelForCausalLM": "modeling_deepseek.DeepseekForCausalLM" }, "aux_loss_alpha": 0.001, "bos_token_id": 128000, "eos_token_id": 128001, "first_k_dense_replace": 0, "hidden_act": "silu", "hidden_size": 4096, "initializer_range": 0.02, "intermediate_size": 14336, "max_position_embeddings": 8192, "model_type": "deepseek", "moe_intermediate_size": 14336, "moe_layer_freq": 1, "n_routed_experts": 4, "n_shared_experts": 1, "norm_topk_prob": true, "num_attention_heads": 32, "num_experts_per_tok": 4, "num_hidden_layers": 32, "num_key_value_heads": 8, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 500000.0, "scoring_func": "softmax", "seq_aux": true, "tie_word_embeddings": false, "torch_dtype": "float16", "transformers_version": "4.40.0.dev0", "use_cache": true, "vocab_size": 128256 }