README.md

---
dataset_info:
  features:
  - name: url
    dtype: string
  - name: commit
    dtype: string
  - name: file_path
    dtype: string
  - name: full_name
    dtype: string
  - name: start
    dtype: string
  - name: end
    dtype: string
  - name: tactic
    dtype: string
  - name: state_before
    dtype: string
  - name: state_after
    dtype: string
  - name: input
    dtype: string
  splits:
  - name: train
    num_bytes: 950103073
    num_examples: 218866
  download_size: 119669803
  dataset_size: 950103073
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---