config.json

{
  "architectures": [
    "Qwen3_5ForCausalLM"
  ],
  "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,
  "intermediate_size": 6144,
  "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_conv_kernel_dim": 4,
  "linear_key_head_dim": 128,
  "linear_num_key_heads": 16,
  "linear_num_value_heads": 16,
  "linear_value_head_dim": 128,
  "mamba_ssm_dtype": "float32",
  "max_position_embeddings": 262144,
  "mlp_only_layers": [],
  "model_type": "qwen3_5_text",
  "mtp_num_hidden_layers": 1,
  "mtp_use_dedicated_embeddings": false,
  "num_attention_heads": 8,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "pad_token_id": null,
  "partial_rotary_factor": 0.25,
  "quantization_config": {
    "config_groups": {
      "group_0": {
        "format": "dense",
        "input_activations": null,
        "output_activations": null,
        "targets": [
          "Linear"
        ],
        "weights": {
          "actorder": null,
          "block_structure": null,
          "dynamic": false,
          "group_size": 128,
          "num_bits": 4,
          "observer": "memoryless_minmax",
          "observer_kwargs": {},
          "scale_dtype": null,
          "strategy": "group",
          "symmetric": false,
          "type": "int",
          "zp_dtype": "torch.int8"
        }
      }
    },
    "format": "dense",
    "global_compression_ratio": null,
    "ignore": [
      "model.layers.0.linear_attn.in_proj_b",
      "model.layers.0.linear_attn.in_proj_a",
      "model.layers.1.linear_attn.in_proj_b",
      "model.layers.1.linear_attn.in_proj_a",
      "model.layers.2.linear_attn.in_proj_b",
      "model.layers.2.linear_attn.in_proj_a",
      "model.layers.4.linear_attn.in_proj_b",
      "model.layers.4.linear_attn.in_proj_a",
      "model.layers.5.linear_attn.in_proj_b",
      "model.layers.5.linear_attn.in_proj_a",
      "model.layers.6.linear_attn.in_proj_b",
      "model.layers.6.linear_attn.in_proj_a",
      "model.layers.8.linear_attn.in_proj_b",
      "model.layers.8.linear_attn.in_proj_a",
      "model.layers.9.linear_attn.in_proj_b",
      "model.layers.9.linear_attn.in_proj_a",
      "model.layers.10.linear_attn.in_proj_b",
      "model.layers.10.linear_attn.in_proj_a",
      "model.layers.12.linear_attn.in_proj_b",
      "model.layers.12.linear_attn.in_proj_a",
      "model.layers.13.linear_attn.in_proj_b",
      "model.layers.13.linear_attn.in_proj_a",
      "model.layers.14.linear_attn.in_proj_b",
      "model.layers.14.linear_attn.in_proj_a",
      "model.layers.16.linear_attn.in_proj_b",
      "model.layers.16.linear_attn.in_proj_a",
      "model.layers.17.linear_attn.in_proj_b",
      "model.layers.17.linear_attn.in_proj_a",
      "model.layers.18.linear_attn.in_proj_b",
      "model.layers.18.linear_attn.in_proj_a",
      "model.layers.20.linear_attn.in_proj_b",
      "model.layers.20.linear_attn.in_proj_a",
      "model.layers.21.linear_attn.in_proj_b",
      "model.layers.21.linear_attn.in_proj_a",
      "model.layers.22.linear_attn.in_proj_b",
      "model.layers.22.linear_attn.in_proj_a",
      "lm_head"
    ],
    "kv_cache_scheme": null,
    "quant_method": "compressed-tensors",
    "quantization_status": "compressed",
    "sparsity_config": {},
    "transform_config": {},
    "version": "0.14.1.a20260225"
  },
  "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"
  },
  "tie_word_embeddings": true,
  "transformers_version": "5.2.0",
  "use_cache": true,
  "vocab_size": 248320
}