{ "architectures": ["Qwen3ForCausalLM"], "auto_map": { "AutoModel": "modeling_qwen3.Qwen3Model", "AutoModelForCausalLM": "modeling_qwen3.Qwen3ForCausalLM" }, "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151643, "pad_token_id": 151643, "dtype": "bfloat16", "head_dim": 128, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 6144, "max_position_embeddings": 131072, "max_window_layers": 24, "mix_coeff": 0, "model_type": "qwen3", "num_attention_heads": 16, "num_hidden_layers": 24, "num_key_value_heads": 16, "rms_norm_eps": 1e-6, "rope_parameters": {"rope_theta": 1000000, "rope_type": "default"}, "sliding_window": null, "summary_chunk_position_ids_type": "origin", "summary_chunk_size": 8, "summary_independent_attention_layernorm": false, "summary_independent_parameters": true, "summary_sliding_chunk_num": [128,128,128,16384,128,128,128,16384,128,128,128,16384,128,128,128,16384,128,128,128,16384,128,128,128,16384], "summary_token_begin": 151936, "summary_token_num": 1, "summary_token_position_ids_type": "last_chunk_slice_right", "tie_word_embeddings": false, "transformers_version": "5.0.0", "truncate_predict_nums": 151936, "use_cache": true, "use_sliding_window": false, "use_summary_attention": true, "vocab_size": 151937 }