{
  "architectures": [
    "HybridGatedDeltaNetForCausalLM"
  ],
  "attn": {
    "attn_type": "lte",
    "buffer_size": 512,
    "conv_dilation": [
      2,
      2,
      2
    ],
    "conv_size": [
      3,
      3,
      3
    ],
    "dropout_prob": 0.1,
    "l1_alpha": 1.5,
    "l1_alpha_neg": 2,
    "l1_lambda_init": 1e-09,
    "l1_lambda_initial_steps": 0,
    "l1_lambda_max": 1.0,
    "l1_lambda_min": 1e-09,
    "l1_lambda_shared": "head",
    "l1_lambda_tolerate": [
      0.95,
      1.0
    ],
    "l1_lambda_update_steps": 32,
    "l1_warmup_steps": 0,
    "layers": [
      1,
      3,
      5,
      7,
      9,
      11,
      13,
      15,
      17,
      19,
      21,
      23
    ],
    "num_heads": 32,
    "num_kv_heads": 8,
    "qkv_bias": false,
    "rope_theta": 10000.0,
    "routing_freeze_steps": 0,
    "routing_grad": true,
    "routing_grad_mode": "mask_sum",
    "routing_method": "sigmoid",
    "routing_noise_steps": 0,
    "routing_shared": "head",
    "routing_with_rope": false,
    "sink_size": 4,
    "window_size": 767
  },
  "attn_mode": "chunk",
  "bos_token_id": 1,
  "conv_size": 4,
  "eos_token_id": 2,
  "expand_v": 1,
  "fuse_cross_entropy": true,
  "fuse_norm": true,
  "fuse_swiglu": true,
  "head_dim": 256,
  "hidden_act": "swish",
  "hidden_ratio": 4,
  "hidden_size": 2048,
  "initializer_range": 0.006,
  "intermediate_size": null,
  "max_position_embeddings": 4096,
  "model_type": "hybrid_gated_deltanet",
  "norm_eps": 1e-06,
  "num_heads": 8,
  "num_hidden_layers": 24,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_gate": true,
  "use_short_conv": true,
  "vocab_size": 32000
}