Alignment dataset tracked in the /datasets sub-catalog.
1:3 ratio of tv-alpaca-v5 to open-instruct-uncensored-alpaca all at 512 LLaMa tokens or less (-25 or so to account for formatting). I didn't trundicate, I just completely removed the ones larger than 512 cause I didn't want to deal with them.
tulu uncensored: 40,353
tv alpaca v5: 13,217
total: 53,570
1:3 ratio of tv-alpaca-v5 to open-instruct-uncensored-alpaca all at 512 LLaMa tokens or less (-25 or so to account for formatting). I didn't trundicate, I just completely removed the ones larger than 512 cause I didn't want to deal with them.
tulu uncensored: 40,353
tv alpaca v5: 13,217
total: 53,570
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.
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.
Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.
from datasets import load_dataset
ds = load_dataset("xzuyn/tv-alpaca-open-instruct-uncensored-blend")from huggingface_hub import snapshot_download
snapshot_download(repo_id="xzuyn/tv-alpaca-open-instruct-uncensored-blend", repo_type="dataset")git clone https://huggingface.co/datasets/xzuyn/tv-alpaca-open-instruct-uncensored-blendEvery 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.