{ "architectures": [ "Qwen3Model" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 151643, "dtype": "bfloat16", "eos_token_id": 151645, "head_dim": 128, "hidden_act": "silu", "hidden_size": 2560, "initializer_range": 0.02, "intermediate_size": 9728, "layer_types": [ "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention" ], "max_position_embeddings": 40960, "max_window_layers": 36, "model_type": "qwen3", "num_attention_heads": 32, "num_hidden_layers": 36, "num_key_value_heads": 8, "pad_token_id": null, "quantization_config": { "config_groups": { "group_0": { "format": "nvfp4-pack-quantized", "input_activations": { "actorder": null, "block_structure": null, "dynamic": "local", "group_size": 16, "num_bits": 4, "observer": "static_minmax", "observer_kwargs": {}, "scale_dtype": "torch.float8_e4m3fn", "strategy": "tensor_group", "symmetric": true, "type": "float", "zp_dtype": null }, "output_activations": null, "targets": [ "re:.*mlp.gate_proj$", "re:.*mlp.up_proj$", "re:.*mlp.down_proj$" ], "weights": { "actorder": null, "block_structure": null, "dynamic": false, "group_size": 16, "num_bits": 4, "observer": "memoryless_minmax", "observer_kwargs": {}, "scale_dtype": "torch.float8_e4m3fn", "strategy": "tensor_group", "symmetric": true, "type": "float", "zp_dtype": null } } }, "format": "nvfp4-pack-quantized", "global_compression_ratio": null, "ignore": [ "re:.*self_attn.*", "layers.0.self_attn.q_proj", "layers.0.self_attn.k_proj", "layers.0.self_attn.v_proj", "layers.0.self_attn.o_proj", "layers.1.self_attn.q_proj", "layers.1.self_attn.k_proj", "layers.1.self_attn.v_proj", "layers.1.self_attn.o_proj", "layers.2.self_attn.q_proj", "layers.2.self_attn.k_proj", "layers.2.self_attn.v_proj", "layers.2.self_attn.o_proj", "layers.3.self_attn.q_proj", "layers.3.self_attn.k_proj", "layers.3.self_attn.v_proj", "layers.3.self_attn.o_proj", "layers.4.self_attn.q_proj", "layers.4.self_attn.k_proj", "layers.4.self_attn.v_proj", "layers.4.self_attn.o_proj", "layers.5.self_attn.q_proj", "layers.5.self_attn.k_proj", "layers.5.self_attn.v_proj", "layers.5.self_attn.o_proj", "layers.6.self_attn.q_proj", "layers.6.self_attn.k_proj", "layers.6.self_attn.v_proj", "layers.6.self_attn.o_proj", "layers.7.self_attn.q_proj", "layers.7.self_attn.k_proj", "layers.7.self_attn.v_proj", "layers.7.self_attn.o_proj", "layers.8.self_attn.q_proj", "layers.8.self_attn.k_proj", "layers.8.self_attn.v_proj", "layers.8.self_attn.o_proj", "layers.9.self_attn.q_proj", "layers.9.self_attn.k_proj", "layers.9.self_attn.v_proj", "layers.9.self_attn.o_proj", "layers.10.self_attn.q_proj", "layers.10.self_attn.k_proj", "layers.10.self_attn.v_proj", "layers.10.self_attn.o_proj", "layers.11.self_attn.q_proj", "layers.11.self_attn.k_proj", "layers.11.self_attn.v_proj", "layers.11.self_attn.o_proj", "layers.12.self_attn.q_proj", "layers.12.self_attn.k_proj", "layers.12.self_attn.v_proj", "layers.12.self_attn.o_proj", "layers.13.self_attn.q_proj", "layers.13.self_attn.k_proj", "layers.13.self_attn.v_proj", "layers.13.self_attn.o_proj", "layers.14.self_attn.q_proj", "layers.14.self_attn.k_proj", "layers.14.self_attn.v_proj", "layers.14.self_attn.o_proj", "layers.15.self_attn.q_proj", "layers.15.self_attn.k_proj", "layers.15.self_attn.v_proj", "layers.15.self_attn.o_proj", "layers.16.self_attn.q_proj", "layers.16.self_attn.k_proj", "layers.16.self_attn.v_proj", "layers.16.self_attn.o_proj", "layers.17.self_attn.q_proj", "layers.17.self_attn.k_proj", "layers.17.self_attn.v_proj", "layers.17.self_attn.o_proj", "layers.18.self_attn.q_proj", "layers.18.self_attn.k_proj", "layers.18.self_attn.v_proj", "layers.18.self_attn.o_proj", "layers.19.self_attn.q_proj", "layers.19.self_attn.k_proj", "layers.19.self_attn.v_proj", "layers.19.self_attn.o_proj", "layers.20.self_attn.q_proj", "layers.20.self_attn.k_proj", "layers.20.self_attn.v_proj", "layers.20.self_attn.o_proj", "layers.21.self_attn.q_proj", "layers.21.self_attn.k_proj", "layers.21.self_attn.v_proj", "layers.21.self_attn.o_proj", "layers.22.self_attn.q_proj", "layers.22.self_attn.k_proj", "layers.22.self_attn.v_proj", "layers.22.self_attn.o_proj", "layers.23.self_attn.q_proj", "layers.23.self_attn.k_proj", "layers.23.self_attn.v_proj", "layers.23.self_attn.o_proj", "layers.24.self_attn.q_proj", "layers.24.self_attn.k_proj", "layers.24.self_attn.v_proj", "layers.24.self_attn.o_proj", "layers.25.self_attn.q_proj", "layers.25.self_attn.k_proj", "layers.25.self_attn.v_proj", "layers.25.self_attn.o_proj", "layers.26.self_attn.q_proj", "layers.26.self_attn.k_proj", "layers.26.self_attn.v_proj", "layers.26.self_attn.o_proj", "layers.27.self_attn.q_proj", "layers.27.self_attn.k_proj", "layers.27.self_attn.v_proj", "layers.27.self_attn.o_proj", "layers.28.self_attn.q_proj", "layers.28.self_attn.k_proj", "layers.28.self_attn.v_proj", "layers.28.self_attn.o_proj", "layers.29.self_attn.q_proj", "layers.29.self_attn.k_proj", "layers.29.self_attn.v_proj", "layers.29.self_attn.o_proj", "layers.30.self_attn.q_proj", "layers.30.self_attn.k_proj", "layers.30.self_attn.v_proj", "layers.30.self_attn.o_proj", "layers.31.self_attn.q_proj", "layers.31.self_attn.k_proj", "layers.31.self_attn.v_proj", "layers.31.self_attn.o_proj", "layers.32.self_attn.q_proj", "layers.32.self_attn.k_proj", "layers.32.self_attn.v_proj", "layers.32.self_attn.o_proj", "layers.33.self_attn.q_proj", "layers.33.self_attn.k_proj", "layers.33.self_attn.v_proj", "layers.33.self_attn.o_proj", "layers.34.self_attn.q_proj", "layers.34.self_attn.k_proj", "layers.34.self_attn.v_proj", "layers.34.self_attn.o_proj", "layers.35.self_attn.q_proj", "layers.35.self_attn.k_proj", "layers.35.self_attn.v_proj", "layers.35.self_attn.o_proj" ], "kv_cache_scheme": null, "quant_method": "compressed-tensors", "quantization_status": "compressed", "sparsity_config": {}, "transform_config": {}, "version": "0.15.1.dev14+g01a1c9a" }, "rms_norm_eps": 1e-06, "rope_parameters": { "rope_theta": 1000000, "rope_type": "default" }, "sliding_window": null, "tie_word_embeddings": true, "transformers_version": "5.2.0", "use_cache": true, "use_sliding_window": false, "vocab_size": 151665 }