{
  "_attn_implementation_autoset": true,
  "_name_or_path": "SeerAttention/SeerAttention-Qwen2.5-32B-Instruct-AttnGates",
  "architectures": [
    "SeerAttnQwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "base_model": "Qwen/Qwen2.5-32B-Instruct",
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 27648,
  "max_position_embeddings": 32768,
  "max_window_layers": 70,
  "model_type": "qwen2",
  "num_attention_heads": 40,
  "num_hidden_layers": 64,
  "num_key_value_heads": 8,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "seerattn_gate_block_size": 64,
  "seerattn_gate_force_double": false,
  "seerattn_gate_hidden_size": 128,
  "seerattn_gate_type": "Qavg_Kmaxminavg",
  "seerattn_last_block_dense": true,
  "seerattn_nz_ratio": 1.0,
  "seerattn_sparsity_method": "threshold",
  "seerattn_threshold": 0.0,
  "sliding_window": 131072,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}