{
  "architectures": [
    "OuroForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_ouro.OuroConfig",
    "AutoModelForCausalLM": "modeling_ouro.OuroForCausalLM"
  },
  "bos_token_id": 1,
  "early_exit_threshold": null,
  "embedding_size": 32,
  "eos_token_id": 2,
  "ffn_gated": true,
  "hidden_act": "silu",
  "hidden_size": 32,
  "initializer_range": 0.02,
  "input_embedding_injection_coefficient": 0.0,
  "intermediate_size": 48,
  "layer_types": [
    "full_attention"
  ],
  "loop_aux_loss_weight": 0.0,
  "max_position_embeddings": 2048,
  "max_window_layers": 28,
  "model_type": "ouro",
  "num_attention_heads": 2,
  "num_hidden_layers": 1,
  "num_key_value_heads": 1,
  "pad_token_id": 0,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "sliding_window": null,
  "student_architecture_source": "ByteDance/Ouro-1.4B@574fa66cb8bf5abdc979642d01cf2b79b16bfab1",
  "student_loss_mode": "last_loop",
  "tie_word_embeddings": true,
  "torch_dtype": "float32",
  "total_ut_steps": 4,
  "transformers_version": "4.55.0",
  "use_cache": true,
  "use_loop_embeddings": false,
  "use_sliding_window": false,
  "value_residual_mix": 0.5,
  "vocab_size": 378
}