Steerling-8B
Shares rank 50 of 72 language models by recommenders
Language model that shows which learned concept produced each token, aiding debugging. Critics call its claims overstated and its demo too broad.
Language model that shows which learned concept produced each token, aiding debugging. Critics call its claims overstated and its demo too broad.
12 people gave an opinion
- recommend it
- 10
- criticize it
- 2
- approval share
- 83%
Recommended in 1 of 13 months
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 10 people recommend itNeeds 10 or more, met.
- 5 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 1 of 13 monthsNeeds 4 or more, not met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Explains which concept produced each token
“our model does more than this. For any chunk the model generates, it can answer: which concept, in the model's representations, was responsible for that token(s)”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” “The model was actually designed to answer these questions for any sentence”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” Helps debug and explain hallucinations
“this now gives us a huge boost in debugging the model”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” “this can tell you exactly why the model hallucinated”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” Makes the model more transparent
“Always happy to see improvements on explanable LLMs.”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” “makes the black box slightly more transparent”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” Concepts built in during training
“Training from scratch AND with supervision allows you to sidestep these issues, and even learn more disentangled representations.”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates” “We constrained the model to do exactly this during training”
Comment from 24 Feb 2026 in “Show HN: Steerling-8B, a language model that can explain any token it generates”
Other language models people recommend
The four most recommended other language models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Claude Opus | 3,331 | 2,746 | 55% | 13 | 4 mo156 owners | — |
| OpenAI GPT | 1,205 | 987 | 55% | 13 | 6 mo47 owners | — |
| Claude Sonnet | 854 | 586 | 59% | 13 | 6 mo46 owners | — |
| DeepSeek V4 Flash | 828 | 223 | 79% | 8 | 1 mo24 owners | Approved |
| Steerling-8BThis page | 10 | 2 | 83% | 1 | — | — |
What people said
18 of the 19 comments that mention it, newest first. Each links to the original comment.
“something I'd find directly useful in my work”
“I think this is super cool and I’m suddenly really interested in experiments like this”
“The demo just says "Wikipedia" or "ArXiV". That's pretty broad and maybe not that useful.”
“All the claims in the post are supported by evidence, no marketing here.”
“Seems pretty cool. You can simply block the concept of tiananmen square and it will be permanently removed from the brain.”
“Always happy to see improvements on explanable LLMs.”
“Looks neat and original, congrats!”
“this is actually the first one that i think has a very serious potential”
“This seems really interesting.”
“this can tell you exactly why the model hallucinated”
“this now gives us a huge boost in debugging the model”
“Training from scratch AND with supervision allows you to sidestep these issues, and even learn more disentangled representations.”
“We constrained the model to do exactly this during training”
“The model was actually designed to answer these questions for any sentence”
“We can find the training data that is causing a model to output toxic/unwanted text and correct it.”
“our model does more than this. For any chunk the model generates, it can answer: which concept, in the model's representations, was responsible for that token(s)”
“Either I'm missing something or this is way overstated.”
“makes the black box slightly more transparent”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.