{ "architectures": [ "Qwen3ForCausalLM" ], "auto_map": { "AutoConfig": "configuration_qwen3.Qwen3Config", "AutoModel": "modeling_qwen3.Qwen3Model", "AutoModelForCausalLM": "modeling_qwen3.Qwen3ForCausalLM" }, "attention_dropout": 0.0, "elementwise_attn_output_gate": false, "headwise_attn_output_gate": true, "bos_token_id": 151643, "eos_token_id": 151643, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 6144, "max_position_embeddings": 32768, "max_window_layers": 28, "model_type": "qwen2", "num_attention_heads": 16, "num_hidden_layers": 28, "num_key_value_heads": 8, "qkv_bias": false, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 1000000, "sliding_window": null, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.46.0", "use_cache": true, "use_qk_norm": true, "use_sliding_window": false, "vocab_size": 152064 }