README.md

---
dataset_info:
  features:
  - name: conversation_hash
    dtype: string
  - name: model
    dtype: string
  - name: dataset
    list: string
  - name: ground_truth
    list: string
  - name: predicted_label
    dtype: string
  - name: custom_id
    dtype: string
  - name: original_dataset
    dtype: string
  - name: dataset_source
    dtype: string
  - name: input_ids_prompt
    list: int64
  - name: input_ids
    list: int32
  - name: attention_mask
    list: int8
  - name: labels
    list: int64
  - name: prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 126731828
    num_examples: 12841
  download_size: 32951909
  dataset_size: 126731828
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---