Get the app
evaluate 0 just added 0

submarat/mats11-mix4-readout2x

published by @submarat

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
102
Last 30 days
102
Likes
-
Size
14.0 GB
Created on HF
2026-10-14
Age
7 days ago

Description

Artifacts of sandbox/marat/2026-10-12-ge-mix4-readout2x of github.com/submarat/mats11-apollo-science-of-scheming:

OLMo-3-1025-7B base, RL on GSM8K/MATH/IFEval, LoRA rank 8. Paths below are relative to that experiment folder.

runs/<run>/measurements/rows/step_XXXX_readout.jsonl.gz: the 768 readout completions at checkpoint t

runs/<run>/logs/exp_001/train_data_step{t+1}.jsonl: update t's 96 training completions with rewards

runs/<run>/measurements/compressed_gradients/step_XXXX/*.pt: bfloat16… See the full description on the dataset page: https://huggingface.co/datasets/submarat/mats11-mix4-readout2x.

Tags

license:othersize_categories:1K<n<10Kformat:jsonmodality:textlibrary:datasetslibrary:dasklibrary:polarslibrary:mlcroissantregion:us

README current version from Hugging Face

pretty_name: mix4 on NeMo (LoRA rank 8, 96 readout prompts) - gradient sketches and rollouts
license: other

Artifacts of sandbox/marat/2026-10-12-ge-mix4-readout2x of github.com/submarat/mats11-apollo-science-of-scheming:
OLMo-3-1025-7B base, RL on GSM8K/MATH/IFEval, LoRA rank 8. Paths below are relative to that experiment folder.

  • runs/<run>/measurements/rows/step_XXXX_readout.jsonl.gz: the 768 readout completions at checkpoint t
  • runs/<run>/logs/exp_001/train_data_step{t+1}.jsonl: update t's 96 training completions with rewards
  • runs/<run>/measurements/compressed_gradients/step_XXXX/*.pt: bfloat16 [rows, 262144] SRHT sketches (seed 0) of per-rollout,
    behavior, loss and update gradients
  • runs/<run>/logs/...: W&B and tensorboard logs

Fetch: python tools/artifacts.py pull <experiment folder> (the committed artifacts.json points at this dataset); no token is needed.
The completions are base-model samples on public benchmark prompts.

README history 7 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.

  1. 2026-10-14copy 52 files78c2fdc977 B
    Loading...
  2. 2026-10-14copy 42 files25a6450977 B
    Loading...
  3. 2026-10-14copy 42 filesba846ae977 B
    Loading...
  4. 2026-10-14copy 42 filesf743f72977 B
    Loading...
  5. 2026-10-14copy 67 filescde3e4e977 B
    Loading...
  6. 2026-10-14dataset card5e951ed977 B
    Loading...
  7. 2026-10-14initial commit5747fd70 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 1 thread · 1 open · 1 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.

  1. 2026-10-15[bot] Conversion to Parquetopen1 💬#1
    Loading...

Continue on Hugging Face →

Related datasets same stage · evaluate

google/IFEval
1.7M dl 182 likes 210 KB
Dataset Card for IFEval Dataset Summary This dataset contains the prompts used in the Instruction-Following Ev...
JailbreakBench/JBB-Behaviors
418.2K dl 143 likes 992 KB
An Open Robustness Benchmark for Jailbreaking Language Models NeurIPS 2024 Datasets and Benchmarks Track P...
Beijing-AISI/panda-bench
195.6K dl 3.1 GB
PandaBench PandaBench is a comprehensive benchmark for evaluating Large Language Model (LLM) safety, focusing on jailbreak atta...
allenai/wildguardmix
137.3K dl 108 likes 53.4 MB
Dataset Card for WildGuardMix Disclaimer: The data includes examples that might be disturbing, harmful or upsetting....
walledai/HarmBench
135.7K dl 59 likes 109 KB
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal Paper: HarmBench: A Standardized Ev...
walledai/XSTest
108.3K dl 30 likes 22 KB
XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models Paper: XSTest: A Test Suite for Ide...
bench-llm/or-bench
77.2K dl 22 likes 32.1 MB
OR-Bench: An Over-Refusal Benchmark for Large Language Models Please see our demo at HuggingFace Spaces. Overall P...
argilla/magpie-ultra-v1.0
65.5K dl 53 likes 20.7 GB
Dataset Card for magpie-ultra-v1.0 This dataset has been created with distilabel. Dataset Summary magpie...

Read further

Community best practices (evaluation)
4 rules on judges, benchmarks, and reporting.
Reference-set drift over time
Why the same benchmark means different things across model generations.
How we label the method for each model
The 12 classification rules that assign M1-M8 to every model.
Non-English abliteration communities
The Korean scepticism about one-click tools and non-English benchmark blindspots.

Metadata

License
other

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("submarat/mats11-mix4-readout2x")
Raw snapshot (Python) 
from huggingface_hub import snapshot_download
snapshot_download(repo_id="submarat/mats11-mix4-readout2x", repo_type="dataset")
git clone (requires git-lfs) 
git clone https://huggingface.co/datasets/submarat/mats11-mix4-readout2x
Browse repository on Hugging Face ->
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration