{
  "_name_or_path": "/cm/archive/namlh35/SATD/results/my-dataset-0.0.1/deepseek-coder-1.3b-multilabel_kfold_code-0.0.2/0",
  "architectures": [
    "LlamaForSequenceClassification"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 32013,
  "eos_token_id": 32014,
  "finetuning_task": "text-classification",
  "hidden_act": "silu",
  "hidden_size": 2048,
  "id2label": {
    "0": "LABEL_0",
    "1": "LABEL_1",
    "2": "LABEL_2",
    "3": "LABEL_3"
  },
  "initializer_range": 0.02,
  "intermediate_size": 5504,
  "label2id": {
    "0": 0,
    "1": 1
  },
  "max_position_embeddings": 16384,
  "max_query_length": 512,
  "max_seq_length": -4,
  "model_type": "llama",
  "num_attention_heads": 16,
  "num_hidden_layers": 24,
  "num_key_value_heads": 16,
  "pad_token_id": 32014,
  "pretraining_tp": 1,
  "problem_type": "multi_label_classification",
  "rms_norm_eps": 1e-06,
  "rope_scaling": {
    "factor": 4.0,
    "type": "linear"
  },
  "rope_theta": 100000,
  "sc_comment": false,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.40.2",
  "use_cache": true,
  "vocab_size": 32256
}