Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Datasets:
KikoCis
/
finetuning-on-the-benchmark-hurts

Modalities:
Text
Formats:
json
Languages:
English
Size:
< 1K
Tags:
evaluation
methodology
fine-tuning
distillation
gsm8k
reasoning
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files Files and versions
xet
Community
1
finetuning-on-the-benchmark-hurts / data
163 kB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 1 commit
KikoCis's picture
KikoCis
Fine-tuning on the benchmark hurts: 3-way study (base/naive/distilled) + fair eval + examples
6e0051b verified 4 months ago
  • examples.jsonl
    13.5 kB
    Fine-tuning on the benchmark hurts: 3-way study (base/naive/distilled) + fair eval + examples 4 months ago
  • gsm8k_test_200.jsonl
    110 kB
    Fine-tuning on the benchmark hurts: 3-way study (base/naive/distilled) + fair eval + examples 4 months ago
  • results.json
    1.96 kB
    Fine-tuning on the benchmark hurts: 3-way study (base/naive/distilled) + fair eval + examples 4 months ago
  • svamp_test_200.jsonl
    37.3 kB
    Fine-tuning on the benchmark hurts: 3-way study (base/naive/distilled) + fair eval + examples 4 months ago