---
dataset_info:
  features:
  - name: Prompt
    dtype: string
  - name: LLM1
    dtype: string
  - name: LLM2
    dtype: string
  - name: LLM1 response
    dtype: string
  - name: LLM2 response
    dtype: string
  - name: Human judges quality
    dtype: string
  - name: Human judges correctness
    dtype: string
  - name: Human judges relevance
    dtype: string
  - name: Human LLM1 Tunisian usage score
    dtype: float64
  - name: Human LLM2 Tunisian usage score
    dtype: float64
  - name: Human LLM1 flagged issues
    dtype: string
  - name: Human LLM2 flagged issues
    dtype: string
  - name: Evaluated
    dtype: bool
  - name: CurrentPromptIndex
    dtype: int64
  splits:
  - name: train
    num_bytes: 1781579
    num_examples: 900
  download_size: 876328
  dataset_size: 1781579
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---