{"slug": "neuronpedia", "name": "Neuronpedia", "category": "ai", "category_name": "AI models and assistants", "description": "", "recommenders": 3, "commenters": 3, "warners": 0, "score": 2.721, "rising": -0.98, "first": "2026-02-22", "latest": "2026-07-21", "why": "there is also https://www.neuronpedia.org/ if one is interested in seeing interactive demonstrations", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 0, "compare": 0, "warn": 0}, "spark": [0, 0, 0, 1, 0, 0, 0, 0, 2, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "there is also https://www.neuronpedia.org/ if one is interested in seeing interactive demonstrations", "kind": "", "icon": true, "domain": "neuronpedia.org", "github_repo": null, "official_url": "https://www.neuronpedia.org", "pricing": "", "aliases": ["Neuronpedia"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 783911153, "item_id": 48994447, "url": "https://news.ycombinator.com/item?id=48994447", "by": "wyrdcurt", "date": "2026-07-21", "stance": "recommend", "quote": "there is also https://www.neuronpedia.org/ if one is interested in seeing interactive demonstrations", "excerpt": "I'm talking about using mechanistic interpretability to see the model's intent. If it is deliberately using compromised libraries to weaken some code's security, there's going to be a signal in its hidden activations that it's doing so. Finding these kinds of activations is something Anthropic is actively researching [1] but they're the only ones who can use those techniques to see Claude's intent. On the other hand,", "context": "", "use_case": "Interpretability demos", "switched_from": "", "story": "Who's Afraid of Chinese Models?"}, {"id": 781663552, "item_id": 48853972, "url": "https://news.ycombinator.com/item?id=48853972", "by": "ada1981", "date": "2026-07-09", "stance": "recommend", "quote": "There\u2019s is likely better!", "excerpt": "I actually just saw the paper not the one they made. Just googled and found it. There\u2019s is likely better! https://www.neuronpedia.org/qwen3.6-27b/jlens?shareId=cmr2kx72r000mpt2x71l23e0z", "context": "", "use_case": "Interpretability demo hosting", "switched_from": "", "story": null}, {"id": 753795377, "item_id": 47112211, "url": "https://news.ycombinator.com/item?id=47112211", "by": "BloondAndDoom", "date": "2026-02-22", "stance": "recommend", "quote": "Possibly can be confirmed something with tools this : https://www.neuronpedia.org/", "excerpt": "Why do you think that? Given how the attention and optimization works on training and inference it makes sense that these kind of words trigger deeper analysis (more steps, introducing more thinking/reasoning steps which wield indeed yield less problems. Even if you make model to spend more time on token outputting you will have more opportunity to emerge better reasoning in between. At least this is how I understand", "context": "", "use_case": "Model interpretability", "switched_from": "", "story": "How I use Claude Code: Separation of planning and execution"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["interpretability demos", "interpretability demo hosting", "model interpretability"], "compared": [], "makers": [{"id": 768921584, "item_id": 48057599, "url": "https://news.ycombinator.com/item?id=48057599", "by": "hijohnnylin", "date": "2026-05-07", "stance": "neutral", "quote": "im from neuronpedia", "excerpt": "hey Nitpicklawyer - Thank you for taking the time to try this out! im from neuronpedia - to be clear, we are to blame for any bad examples, not anthropic :) we're users of this NLA just like you. also, I don't speak for anthropic or the researchers. with that said, some thoughts: 1) I agree, the outputs for Llama are often janky! And I think that might be part of the reason to release this so that people can help ref", "context": "", "use_case": "NLA exploration frontend", "switched_from": "", "story": "Natural Language Autoencoders: Turning Claude's Thoughts into Text"}], "threads": [{"url": "https://news.ycombinator.com/item?id=48977128", "title": "Who's Afraid of Chinese Models?", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48849618", "title": "item 48849618", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48052537", "title": "Natural Language Autoencoders: Turning Claude's Thoughts into Text", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47106686", "title": "How I use Claude Code: Separation of planning and execution", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {"recommend": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {}, "2026-07": {"recommend": 2}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 783911153, "item_id": 48994447, "url": "https://news.ycombinator.com/item?id=48994447", "by": "wyrdcurt", "date": "2026-07-21", "stance": "recommend", "quote": "there is also https://www.neuronpedia.org/ if one is interested in seeing interactive demonstrations", "excerpt": "I'm talking about using mechanistic interpretability to see the model's intent. If it is deliberately using compromised libraries to weaken some code's security, there's going to be a signal in its hidden activations that it's doing so. Finding these kinds of activations is something Anthropic is actively researching [1] but they're the only ones who can use those techniques to see Claude's intent. On the other hand,", "context": "", "use_case": "Interpretability demos", "switched_from": "", "story": "Who's Afraid of Chinese Models?"}, {"id": 781663552, "item_id": 48853972, "url": "https://news.ycombinator.com/item?id=48853972", "by": "ada1981", "date": "2026-07-09", "stance": "recommend", "quote": "There\u2019s is likely better!", "excerpt": "I actually just saw the paper not the one they made. Just googled and found it. There\u2019s is likely better! https://www.neuronpedia.org/qwen3.6-27b/jlens?shareId=cmr2kx72r000mpt2x71l23e0z", "context": "", "use_case": "Interpretability demo hosting", "switched_from": "", "story": null}, {"id": 753795377, "item_id": 47112211, "url": "https://news.ycombinator.com/item?id=47112211", "by": "BloondAndDoom", "date": "2026-02-22", "stance": "recommend", "quote": "Possibly can be confirmed something with tools this : https://www.neuronpedia.org/", "excerpt": "Why do you think that? Given how the attention and optimization works on training and inference it makes sense that these kind of words trigger deeper analysis (more steps, introducing more thinking/reasoning steps which wield indeed yield less problems. Even if you make model to spend more time on token outputting you will have more opportunity to emerge better reasoning in between. At least this is how I understand", "context": "", "use_case": "Model interpretability", "switched_from": "", "story": "How I use Claude Code: Separation of planning and execution"}]}