{
  "architectures": [
    "Step3p5ForCausalLM"
  ],
  "att_impl_type": "GQA",
  "attention_other_setting": {
    "attention_type": "sliding_attention",
    "head_dim": 128,
    "num_attention_groups": 8,
    "num_attention_heads": 96,
    "true_head_dim": 128
  },
  "auto_map": {
    "AutoConfig": "configuration_step3p5.Step3p5Config",
    "AutoModelForCausalLM": "modeling_step3p5.Step3p5ForCausalLM"
  },
  "bos_token_id": 0,
  "dtype": "bfloat16",
  "eos_token_id": [
    1,
    2,
    128007
  ],
  "head_dim": 128,
  "hidden_size": 4096,
  "intermediate_size": 11264,
  "layer_types": [
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention"
  ],
  "max_position_embeddings": 262144,
  "max_seq_len": 262144,
  "model_type": "step3p5",
  "moe_every_n_layer": 1,
  "moe_intermediate_size": 1280,
  "moe_layer_offset": 0,
  "moe_layers_enum": "3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44",
  "moe_num_experts": 288,
  "moe_router_activation": "sigmoid",
  "moe_router_scaling_factor": 3.0,
  "moe_top_k": 8,
  "need_fp32_gate": true,
  "norm_expert_weight": true,
  "num_attention_groups": 8,
  "num_attention_heads": 64,
  "num_hidden_layers": 45,
  "num_nextn_predict_layers": 3,
  "partial_rotary_factor": 0.5,
  "partial_rotary_factors": [
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0,
    0.5,
    1.0,
    1.0,
    1.0
  ],
  "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": [
          "Linear",
          "MoELinear"
        ],
        "weights": {
          "actorder": null,
          "block_structure": null,
          "dynamic": false,
          "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
        }
      }
    },
    "format": "nvfp4-pack-quantized",
    "global_compression_ratio": null,
    "ignore": [
      "lm_head"
    ],
    "kv_cache_scheme": null,
    "quant_method": "compressed-tensors",
    "quantization_status": "compressed",
    "sparsity_config": {},
    "transform_config": {},
    "version": "0.13.0"
  },
  "rms_norm_eps": 1e-05,
  "rope_parameters": {
    "factor": 2.0,
    "high_freq_factor": 32.0,
    "low_freq_factor": 1.0,
    "original_max_position_embeddings": 131072,
    "rope_type": "llama3"
  },
  "rope_scaling": {
    "factor": 2.0,
    "high_freq_factor": 32.0,
    "low_freq_factor": 1.0,
    "original_max_position_embeddings": 131072,
    "rope_type": "llama3"
  },
  "rope_theta": [
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0,
    5000000.0,
    10000.0,
    10000.0,
    10000.0
  ],
  "share_expert_dim": 1280,
  "sink": false,
  "sliding_window": 512,
  "swiglu_limits": [
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    7,
    7,
    0.0,
    0.0,
    0.0
  ],
  "swiglu_limits_shared": [
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    0.0,
    16,
    0.0,
    0.0,
    0.0
  ],
  "transformers_version": "4.57.3",
  "use_cache": false,
  "use_head_wise_attn_gate": true,
  "use_moe": true,
  "use_moe_router_bias": true,
  "use_qk_norm": true,
  "use_rope_layers": [],
  "vocab_size": 128896,
  "yarn_only_types": [
    "full_attention"
  ],
  "zero_centered": true
}