ablate 0 0 0
open-llm-leaderboard-old/details_failspy__Llama-3-8B-Instruct-abliterated
published by @open-llm-leaderboard-old
Alignment dataset tracked in the /datasets sub-catalog.
Lifetime downloads
2.9K
Last 30 days
60
Likes
-
Size
169.6 MB
Created on HF
2024-05-22
Age
2 years ago
Tags
README history
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.
2024-05-23Upload folder using huggingface_hub
Loading...2024-05-23Upload results_2024-05-23T01-26-55.324351.parquet with huggingface_hub
Loading...2024-05-23Upload results_2024-05-23T01-26-55.324351.json with huggingface_hub
Loading...2024-05-23Upload folder using huggingface_hub
Loading...2024-05-23Upload results_2024-05-23T01-24-11.488973.parquet with huggingface_hub
Loading...2024-05-23Upload results_2024-05-23T01-24-11.488973.json with huggingface_hub
Loading...2024-05-23initial commit
Loading...
Used by abliterated models
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.
Related datasets
Read further
Community best practices (ablation)
5 rules from KL-optimisation to layer-0 preservation.
M1 - Direct removal
The canonical Arditi-style ablation.
M3 - Layer-wise ablation
Selective abliteration limited to a band of layers.
Non-English abliteration communities
The Chinese layer-window heuristic and huihui-ai practice.
Repository
Download the dataset
0Copy 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("open-llm-leaderboard-old/details_failspy__Llama-3-8B-Instruct-abliterated") Raw snapshot (Python)
from huggingface_hub import snapshot_download
snapshot_download(repo_id="open-llm-leaderboard-old/details_failspy__Llama-3-8B-Instruct-abliterated", repo_type="dataset") git clone (requires git-lfs)
git clone https://huggingface.co/datasets/open-llm-leaderboard-old/details_failspy__Llama-3-8B-Instruct-abliterated