README.md

metadata

dataset_info:
  features:
    - name: instruction
      dtype: string
    - name: input
      dtype: string
    - name: output
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 44573220
      num_examples: 49622
  download_size: 23365210
  dataset_size: 44573220
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
size_categories:
  - 10K<n<100K
license: cc-by-nc-4.0
language:
  - en

Derived from datasets/tatsu-lab/alpaca.

Deduped with spaces/minishlab/semantic-deduplication, applying a 0.9 similarity threshold to the instruction column The intention was to keep instructions distinct so that trained attention won't overfit to a prompt pattern.