kfkas's picture
Update README.md
6c4fd3a verified
metadata
dataset_info:
  features:
    - name: Instruction
      dtype: string
    - name: word_count
      dtype: int64
    - name: output
      dtype: string
    - name: mlt
      dtype: string
    - name: id
      dtype: int64
  splits:
    - name: train
      num_bytes: 166324683
      num_examples: 121229
  download_size: 97259488
  dataset_size: 166324683
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Ruler Upper Bound Dataset

This dataset contains samples with MLT (Maximum Length Token) upper bound labels.

MLT Distribution

MLT Label Count
[MLT:10] 11316
[MLT:150] 25103
[MLT:300] 13668
[MLT:30] 19608
[MLT:500] 4627
[MLT:50] 18039
[MLT:700] 857
[MLT:80] 19933
[MLT:>800] 8078

Generated automatically during dataset upload.