| { |
| "architectures": [ |
| "Qwen3_5MoeForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attn_output_gate": true, |
| "bos_token_id": null, |
| "dtype": "bfloat16", |
| "eos_token_id": 248044, |
| "full_attention_interval": 4, |
| "head_dim": 256, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "layer_types": [ |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention" |
| ], |
| "linear_conv_kernel_dim": 4, |
| "linear_key_head_dim": 128, |
| "linear_num_key_heads": 16, |
| "linear_num_value_heads": 32, |
| "linear_value_head_dim": 128, |
| "mamba_ssm_dtype": "float32", |
| "max_position_embeddings": 262144, |
| "mlp_only_layers": [], |
| "model_type": "qwen3_5_moe_text", |
| "moe_intermediate_size": 512, |
| "mtp_num_hidden_layers": 1, |
| "mtp_use_dedicated_embeddings": false, |
| "num_attention_heads": 16, |
| "num_experts": 256, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 40, |
| "num_key_value_heads": 2, |
| "output_router_logits": false, |
| "pad_token_id": null, |
| "partial_rotary_factor": 0.25, |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "format": "float-quantized", |
| "input_activations": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": true, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": null, |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "token", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| }, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "memoryless_minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "channel", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| } |
| } |
| }, |
| "format": "float-quantized", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "model.layers.0.linear_attn.out_proj", |
| "model.layers.0.linear_attn.in_proj_qkv", |
| "model.layers.0.linear_attn.in_proj_z", |
| "model.layers.0.linear_attn.in_proj_b", |
| "model.layers.0.linear_attn.in_proj_a", |
| "model.layers.0.mlp.shared_expert_gate", |
| "model.layers.1.linear_attn.out_proj", |
| "model.layers.1.linear_attn.in_proj_qkv", |
| "model.layers.1.linear_attn.in_proj_z", |
| "model.layers.1.linear_attn.in_proj_b", |
| "model.layers.1.linear_attn.in_proj_a", |
| "model.layers.1.mlp.shared_expert_gate", |
| "model.layers.2.linear_attn.out_proj", |
| "model.layers.2.linear_attn.in_proj_qkv", |
| "model.layers.2.linear_attn.in_proj_z", |
| "model.layers.2.linear_attn.in_proj_b", |
| "model.layers.2.linear_attn.in_proj_a", |
| "model.layers.2.mlp.shared_expert_gate", |
| "model.layers.3.mlp.shared_expert_gate", |
| "model.layers.4.linear_attn.out_proj", |
| "model.layers.4.linear_attn.in_proj_qkv", |
| "model.layers.4.linear_attn.in_proj_z", |
| "model.layers.4.linear_attn.in_proj_b", |
| "model.layers.4.linear_attn.in_proj_a", |
| "model.layers.4.mlp.shared_expert_gate", |
| "model.layers.5.linear_attn.out_proj", |
| "model.layers.5.linear_attn.in_proj_qkv", |
| "model.layers.5.linear_attn.in_proj_z", |
| "model.layers.5.linear_attn.in_proj_b", |
| "model.layers.5.linear_attn.in_proj_a", |
| "model.layers.5.mlp.shared_expert_gate", |
| "model.layers.6.linear_attn.out_proj", |
| "model.layers.6.linear_attn.in_proj_qkv", |
| "model.layers.6.linear_attn.in_proj_z", |
| "model.layers.6.linear_attn.in_proj_b", |
| "model.layers.6.linear_attn.in_proj_a", |
| "model.layers.6.mlp.shared_expert_gate", |
| "model.layers.7.mlp.shared_expert_gate", |
| "model.layers.8.linear_attn.out_proj", |
| "model.layers.8.linear_attn.in_proj_qkv", |
| "model.layers.8.linear_attn.in_proj_z", |
| "model.layers.8.linear_attn.in_proj_b", |
| "model.layers.8.linear_attn.in_proj_a", |
| "model.layers.8.mlp.shared_expert_gate", |
| "model.layers.9.linear_attn.out_proj", |
| "model.layers.9.linear_attn.in_proj_qkv", |
| "model.layers.9.linear_attn.in_proj_z", |
| "model.layers.9.linear_attn.in_proj_b", |
| "model.layers.9.linear_attn.in_proj_a", |
| "model.layers.9.mlp.shared_expert_gate", |
| "model.layers.10.linear_attn.out_proj", |
| "model.layers.10.linear_attn.in_proj_qkv", |
| "model.layers.10.linear_attn.in_proj_z", |
| "model.layers.10.linear_attn.in_proj_b", |
| "model.layers.10.linear_attn.in_proj_a", |
| "model.layers.10.mlp.shared_expert_gate", |
| "model.layers.11.mlp.shared_expert_gate", |
| "model.layers.12.linear_attn.out_proj", |
| "model.layers.12.linear_attn.in_proj_qkv", |
| "model.layers.12.linear_attn.in_proj_z", |
| "model.layers.12.linear_attn.in_proj_b", |
| "model.layers.12.linear_attn.in_proj_a", |
| "model.layers.12.mlp.shared_expert_gate", |
| "model.layers.13.linear_attn.out_proj", |
| "model.layers.13.linear_attn.in_proj_qkv", |
| "model.layers.13.linear_attn.in_proj_z", |
| "model.layers.13.linear_attn.in_proj_b", |
| "model.layers.13.linear_attn.in_proj_a", |
| "model.layers.13.mlp.shared_expert_gate", |
| "model.layers.14.linear_attn.out_proj", |
| "model.layers.14.linear_attn.in_proj_qkv", |
| "model.layers.14.linear_attn.in_proj_z", |
| "model.layers.14.linear_attn.in_proj_b", |
| "model.layers.14.linear_attn.in_proj_a", |
| "model.layers.14.mlp.shared_expert_gate", |
| "model.layers.15.mlp.shared_expert_gate", |
| "model.layers.16.linear_attn.out_proj", |
| "model.layers.16.linear_attn.in_proj_qkv", |
| "model.layers.16.linear_attn.in_proj_z", |
| "model.layers.16.linear_attn.in_proj_b", |
| "model.layers.16.linear_attn.in_proj_a", |
| "model.layers.16.mlp.shared_expert_gate", |
| "model.layers.17.linear_attn.out_proj", |
| "model.layers.17.linear_attn.in_proj_qkv", |
| "model.layers.17.linear_attn.in_proj_z", |
| "model.layers.17.linear_attn.in_proj_b", |
| "model.layers.17.linear_attn.in_proj_a", |
| "model.layers.17.mlp.shared_expert_gate", |
| "model.layers.18.linear_attn.out_proj", |
| "model.layers.18.linear_attn.in_proj_qkv", |
| "model.layers.18.linear_attn.in_proj_z", |
| "model.layers.18.linear_attn.in_proj_b", |
| "model.layers.18.linear_attn.in_proj_a", |
| "model.layers.18.mlp.shared_expert_gate", |
| "model.layers.19.mlp.shared_expert_gate", |
| "model.layers.20.linear_attn.out_proj", |
| "model.layers.20.linear_attn.in_proj_qkv", |
| "model.layers.20.linear_attn.in_proj_z", |
| "model.layers.20.linear_attn.in_proj_b", |
| "model.layers.20.linear_attn.in_proj_a", |
| "model.layers.20.mlp.shared_expert_gate", |
| "model.layers.21.linear_attn.out_proj", |
| "model.layers.21.linear_attn.in_proj_qkv", |
| "model.layers.21.linear_attn.in_proj_z", |
| "model.layers.21.linear_attn.in_proj_b", |
| "model.layers.21.linear_attn.in_proj_a", |
| "model.layers.21.mlp.shared_expert_gate", |
| "model.layers.22.linear_attn.out_proj", |
| "model.layers.22.linear_attn.in_proj_qkv", |
| "model.layers.22.linear_attn.in_proj_z", |
| "model.layers.22.linear_attn.in_proj_b", |
| "model.layers.22.linear_attn.in_proj_a", |
| "model.layers.22.mlp.shared_expert_gate", |
| "model.layers.23.mlp.shared_expert_gate", |
| "model.layers.24.linear_attn.out_proj", |
| "model.layers.24.linear_attn.in_proj_qkv", |
| "model.layers.24.linear_attn.in_proj_z", |
| "model.layers.24.linear_attn.in_proj_b", |
| "model.layers.24.linear_attn.in_proj_a", |
| "model.layers.24.mlp.shared_expert_gate", |
| "model.layers.25.linear_attn.out_proj", |
| "model.layers.25.linear_attn.in_proj_qkv", |
| "model.layers.25.linear_attn.in_proj_z", |
| "model.layers.25.linear_attn.in_proj_b", |
| "model.layers.25.linear_attn.in_proj_a", |
| "model.layers.25.mlp.shared_expert_gate", |
| "model.layers.26.linear_attn.out_proj", |
| "model.layers.26.linear_attn.in_proj_qkv", |
| "model.layers.26.linear_attn.in_proj_z", |
| "model.layers.26.linear_attn.in_proj_b", |
| "model.layers.26.linear_attn.in_proj_a", |
| "model.layers.26.mlp.shared_expert_gate", |
| "model.layers.27.mlp.shared_expert_gate", |
| "model.layers.28.linear_attn.out_proj", |
| "model.layers.28.linear_attn.in_proj_qkv", |
| "model.layers.28.linear_attn.in_proj_z", |
| "model.layers.28.linear_attn.in_proj_b", |
| "model.layers.28.linear_attn.in_proj_a", |
| "model.layers.28.mlp.shared_expert_gate", |
| "model.layers.29.linear_attn.out_proj", |
| "model.layers.29.linear_attn.in_proj_qkv", |
| "model.layers.29.linear_attn.in_proj_z", |
| "model.layers.29.linear_attn.in_proj_b", |
| "model.layers.29.linear_attn.in_proj_a", |
| "model.layers.29.mlp.shared_expert_gate", |
| "model.layers.30.linear_attn.out_proj", |
| "model.layers.30.linear_attn.in_proj_qkv", |
| "model.layers.30.linear_attn.in_proj_z", |
| "model.layers.30.linear_attn.in_proj_b", |
| "model.layers.30.linear_attn.in_proj_a", |
| "model.layers.30.mlp.shared_expert_gate", |
| "model.layers.31.mlp.shared_expert_gate", |
| "model.layers.32.linear_attn.out_proj", |
| "model.layers.32.linear_attn.in_proj_qkv", |
| "model.layers.32.linear_attn.in_proj_z", |
| "model.layers.32.linear_attn.in_proj_b", |
| "model.layers.32.linear_attn.in_proj_a", |
| "model.layers.32.mlp.shared_expert_gate", |
| "model.layers.33.linear_attn.out_proj", |
| "model.layers.33.linear_attn.in_proj_qkv", |
| "model.layers.33.linear_attn.in_proj_z", |
| "model.layers.33.linear_attn.in_proj_b", |
| "model.layers.33.linear_attn.in_proj_a", |
| "model.layers.33.mlp.shared_expert_gate", |
| "model.layers.34.linear_attn.out_proj", |
| "model.layers.34.linear_attn.in_proj_qkv", |
| "model.layers.34.linear_attn.in_proj_z", |
| "model.layers.34.linear_attn.in_proj_b", |
| "model.layers.34.linear_attn.in_proj_a", |
| "model.layers.34.mlp.shared_expert_gate", |
| "model.layers.35.mlp.shared_expert_gate", |
| "model.layers.36.linear_attn.out_proj", |
| "model.layers.36.linear_attn.in_proj_qkv", |
| "model.layers.36.linear_attn.in_proj_z", |
| "model.layers.36.linear_attn.in_proj_b", |
| "model.layers.36.linear_attn.in_proj_a", |
| "model.layers.36.mlp.shared_expert_gate", |
| "model.layers.37.linear_attn.out_proj", |
| "model.layers.37.linear_attn.in_proj_qkv", |
| "model.layers.37.linear_attn.in_proj_z", |
| "model.layers.37.linear_attn.in_proj_b", |
| "model.layers.37.linear_attn.in_proj_a", |
| "model.layers.37.mlp.shared_expert_gate", |
| "model.layers.38.linear_attn.out_proj", |
| "model.layers.38.linear_attn.in_proj_qkv", |
| "model.layers.38.linear_attn.in_proj_z", |
| "model.layers.38.linear_attn.in_proj_b", |
| "model.layers.38.linear_attn.in_proj_a", |
| "model.layers.38.mlp.shared_expert_gate", |
| "model.layers.39.mlp.shared_expert_gate", |
| "lm_head" |
| ], |
| "kv_cache_scheme": null, |
| "quant_method": "compressed-tensors", |
| "quantization_status": "compressed", |
| "sparsity_config": {}, |
| "transform_config": {}, |
| "version": "0.14.1.dev3+g928d36b" |
| }, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "mrope_interleaved": true, |
| "mrope_section": [ |
| 11, |
| 11, |
| 10 |
| ], |
| "partial_rotary_factor": 0.25, |
| "rope_theta": 10000000, |
| "rope_type": "default" |
| }, |
| "router_aux_loss_coef": 0.001, |
| "shared_expert_intermediate_size": 512, |
| "tie_word_embeddings": false, |
| "transformers_version": "5.3.0.dev0", |
| "use_cache": true, |
| "vocab_size": 248320 |
| } |
