README.md

metadata

license: mit
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: results
      struct:
        - name: Alpha
          struct:
            - name: mean
              dtype: float64
            - name: ci
              sequence: float64
        - name: Beta
          struct:
            - name: mean
              dtype: float64
            - name: ci
              sequence: float64
        - name: Neutral
          struct:
            - name: mean
              dtype: float64
            - name: ci
              sequence: float64
        - name: Sigma
          struct:
            - name: mean
              dtype: float64
            - name: ci
              sequence: float64
    - name: sigma_score
      dtype: float64
    - name: model_name
      dtype: string
    - name: torch_dtype
      dtype: string
    - name: timestamp
      dtype: string
    - name: n_bootstrap
      dtype: int64
  splits:
    - name: train
      num_bytes: 18677
      num_examples: 95
  download_size: 15158
  dataset_size: 18677
language:
  - en
tags:
  - hellasigma
  - eval
  - personality

pszemraj/hellasigma-results

Results for hellasigma. The sigma score is computed as follows:

To use this as a primitive leaderboard, click on the "sort descending" arrow:

results

oct 30, 2024