Back to datasets
Dataset assetOpen Source CommunityModel EvaluationCommonsense QA

commonsense_qa_eval

The dataset includes multiple features such as question, question concept, options, answer, input, output, etc. Specific fields are id, question, question_concept, choices (with label and text), answerKey, input, extracted_baseline_llama_1b, reasoning_64_a128_mix_mmlu_csqa_gsm8k_even, baseline_llama_1b, output_w_reasoning_llama_1b, extracted_output_w_reasoning_llama_1b, and eval_baseline_vs_mixed_reasoning. It consists of a validation split with 1 221 samples totaling 7 882 991 bytes.

Source
huggingface
Created
Nov 21, 2024
Updated
Dec 5, 2024
Signals
147 views
Availability
Linked source ready
Overview

Dataset description and usage context

Commonsense QA Eval Dataset Overview

Dataset Information

Features

  • id: string
  • question: string
  • question_concept: string
  • choices: sequence
    • label: string
    • text: string
  • answerKey: string
  • input: string
  • extracted_baseline_llama_1b: string
  • reasoning_64_a128_mix_mmlu_csqa_gsm8k_even: string
  • baseline_llama_1b: string
  • output_w_reasoning_llama_1b: string
  • extracted_output_w_reasoning_llama_1b: string
  • eval_baseline_vs_mixed_reasoning: string

Data Split

  • validation:
    • Sample count: 1 221
    • Data size: 7 882 991 bytes

Dataset Size

  • Download size: 3 720 885 bytes
  • Total dataset size: 7 882 991 bytes

Configuration

  • config_name: default
    • data_files:
      • split: validation
      • path: data/validation-*
Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio