{
  "architectures": [
    "Starcoder2ForSequenceClassification"
  ],
  "attention_dropout": 0.0,
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 0,
  "embedding_dropout": 0.0,
  "eos_token_id": 0,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 32,
  "initializer_range": 0.02,
  "intermediate_size": 37,
  "max_position_embeddings": 512,
  "model_type": "starcoder2",
  "norm_epsilon": 1e-05,
  "num_attention_heads": 4,
  "num_hidden_layers": 2,
  "num_key_value_heads": 2,
  "pad_token_id": 0,
  "residual_dropout": 0.0,
  "rope_theta": 10000.0,
  "sliding_window": null,
  "torch_dtype": "float32",
  "transformers_version": "4.40.0.dev0",
  "type_vocab_size": 16,
  "use_bias": true,
  "use_cache": true,
  "vocab_size": 1024
}