README current version from Hugging Face

dataset_info:
features:

name: prompt
dtype: string
name: chosen
dtype: string
name: chosen_model
dtype: string
name: rejected
dtype: string
name: rejected_model
dtype: string
name: subset
dtype: string
name: id
dtype: int64
name: reward_chosen
dtype: float64
name: reward_rejected
dtype: float64
splits:
name: xstest_should_respond
num_bytes: 321610
num_examples: 250
name: hep_go
num_bytes: 179381
num_examples: 164
name: hep_js
num_bytes: 159504
num_examples: 164
name: donotanswer
num_bytes: 293380
num_examples: 136
name: refusals_offensive
num_bytes: 174198
num_examples: 100
name: refusals_dangerous
num_bytes: 275982
num_examples: 100
name: hep_java
num_bytes: 202340
num_examples: 164
name: llmbar_adver_manual
num_bytes: 78234
num_examples: 46
name: xstest_should_refuse
num_bytes: 238780
num_examples: 154
name: alpacaeval_easy
num_bytes: 2169328
num_examples: 805
name: hep_cpp
num_bytes: 168833
num_examples: 164
name: mt_bench_hard
num_bytes: 94718
num_examples: 45
name: llmbar_adver_neighbor
num_bytes: 131026
num_examples: 134
name: mt_bench_med
num_bytes: 127416
num_examples: 45
name: mt_bench_easy
num_bytes: 94345
num_examples: 28
name: llmbar_natural
num_bytes: 90090
num_examples: 100
name: alpacaeval_hard
num_bytes: 1600609
num_examples: 805
name: llmbar_adver_GPTOut
num_bytes: 32156
num_examples: 47
name: hep_rust
num_bytes: 174778
num_examples: 164
name: hep_python
num_bytes: 141969
num_examples: 164
name: alpacaeval_length
num_bytes: 3081227
num_examples: 805
name: math_prm
num_bytes: 878524
num_examples: 447
name: llmbar_adver_GPTInst
num_bytes: 210583
num_examples: 92
download_size: 5981687
dataset_size: 10919011
configs:
config_name: default
data_files:
- split: xstest_should_respond
  path: data/xstest_should_respond-*
- split: hep_go
  path: data/hep_go-*
- split: hep_js
  path: data/hep_js-*
- split: donotanswer
  path: data/donotanswer-*
- split: refusals_offensive
  path: data/refusals_offensive-*
- split: refusals_dangerous
  path: data/refusals_dangerous-*
- split: hep_java
  path: data/hep_java-*
- split: llmbar_adver_manual
  path: data/llmbar_adver_manual-*
- split: xstest_should_refuse
  path: data/xstest_should_refuse-*
- split: alpacaeval_easy
  path: data/alpacaeval_easy-*
- split: hep_cpp
  path: data/hep_cpp-*
- split: mt_bench_hard
  path: data/mt_bench_hard-*
- split: llmbar_adver_neighbor
  path: data/llmbar_adver_neighbor-*
- split: mt_bench_med
  path: data/mt_bench_med-*
- split: mt_bench_easy
  path: data/mt_bench_easy-*
- split: llmbar_natural
  path: data/llmbar_natural-*
- split: alpacaeval_hard
  path: data/alpacaeval_hard-*
- split: llmbar_adver_GPTOut
  path: data/llmbar_adver_GPTOut-*
- split: hep_rust
  path: data/hep_rust-*
- split: hep_python
  path: data/hep_python-*
- split: alpacaeval_length
  path: data/alpacaeval_length-*
- split: math_prm
  path: data/math_prm-*
- split: llmbar_adver_GPTInst
  path: data/llmbar_adver_GPTInst-*

README history 3 revisions

Every night we snapshot the README of every dataset in the catalog. When the SHA changes we archive the new version and diff it against the last. This is the evolving thought record of the alignment-data field: what the author decided to say about the corpus, and how that framing shifted over time.

2025-01-17Upload dataseta8d9f713.3 KB
Loading...
2025-01-17Upload dataset2d6349f3.3 KB
Loading...
2025-01-17initial commitb9ce19a0 B
Loading...

Used by abliterated models none yet

No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.

Community discussions 2 threads · 2 open · 2 comments

Threads posted by the community on Hugging Face: bug reports, questions about usage, pull requests on the README, license clarifications. Click any thread to expand and read the full conversation inline. Newest first.

2025-01-28PRLibrarian Bot: Add language metadata for datasetopen1 💬#2
Loading...
2025-01-18[bot] Conversion to Parquetopen1 💬#1
Loading...

Continue on Hugging Face →

Related datasets same stage · misc

Read further

Community best practices (architecture)

4 rules for MoE, hybrid SSM, FP8 requantization, and Gemma quirks.

The six workflow stages

What misc is doing more work than its name suggests.

Beyond English: multilingual corpora

The four language communities and their workarounds.

Non-English abliteration communities

Deep dive: huihui-ai, layer-window, slop-framing, CJK tokenizer fixes.

Repository

DIR data 23 files FILE .gitattributes FILE README.md

File contents live on Hugging Face. Click a file or folder to open it there.

Download the dataset

0

Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.

Recommended: datasets library (Python)

from datasets import load_dataset
ds = load_dataset("Ayush-Singh/reward-bench-hacking-rewards-harmless-train-normal")

Raw snapshot (Python)

from huggingface_hub import snapshot_download
snapshot_download(repo_id="Ayush-Singh/reward-bench-hacking-rewards-harmless-train-normal", repo_type="dataset")

git clone (requires git-lfs)

git clone https://huggingface.co/datasets/Ayush-Singh/reward-bench-hacking-rewards-harmless-train-normal

Browse repository on Hugging Face ->

Ayush-Singh/reward-bench-hacking-rewards-harmless-train-normal

Tags