{
  "_name_or_path": "/cpfs02/user/jiangyuhua/flash-linear-attention/training/configs/nirvana_1.3B-t3/nirvana_1_3B.json",
  "architectures": [
    "TransformerForCausalLM_rnn"
  ],
  "attention_bias": false,
  "auto_map": {
    "AutoConfig": "configuration_transformer_rnn.TransformerConfig_rnn",
    "AutoModel": "modeling_transformer_rnn.TransformerModel_rnn",
    "AutoModelForCausalLM": "modeling_transformer_rnn.TransformerForCausalLM_rnn"
  },
  "bos_token_id": 1,
  "concept_dim": 64,
  "elementwise_affine": true,
  "eos_token_id": 2,
  "fuse_cross_entropy": true,
  "fuse_norm": true,
  "hidden_act": "swish",
  "hidden_ratio": 4,
  "hidden_size": 2048,
  "initializer_range": 0.006,
  "intermediate_size": null,
  "logit_dim": 32,
  "max_position_embeddings": 32768,
  "model_type": "transformer_rnn",
  "norm_eps": 1e-06,
  "norm_first": false,
  "num_heads": 16,
  "num_hidden_layers": 16,
  "num_kv_heads": null,
  "pad_token_id": 2,
  "recurrent_depth": 4,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.46.0",
  "use_cache": false,
  "vocab_size": 128512,
  "window_size": 2048
}