Hacker GitHub

← All repos

DGoettlich/history-llms

View on GitHub ↗ ★ 1,808

Information hub for our project training the largest possible historical LLMs.

17 mentions on Hacker News by 14 people: 1 story, 16 comments · first 2025-12-18 · latest 2026-10-07

Mentions per year 17 mentions · busiest year 2025 (9)
20082026
AllStoriesComments NewestTopOldest
  1. …LLMs were being trained purely on historical text, have any of them come out yet? Edit: e.g. https://github.com/DGoettlich/history-llms Ten months ago, no update yet... I recall at least one similar project, I'll see if I can find it.
    comment by andai · 2026-10-07 · in Write Like It's 1866: LLMs Relearn Telegraphese
  2. Or https://github.com/DGoettlich/history-llms Discussed at https://news.ycombinator.com/item?id=46319826 (897 points, 431 comments)
    comment by Gander5739 · 2026-04-28 · in Talkie: a 13B vintage language model from 1930
  3. There's a similar but unreleased project here: https://github.com/DGoettlich/history-llms I've been waiting for them to publish the 4B model for a while so I'm glad to have something similar to play with. I think I trust the Ranke-4B…
    comment by aesthesia · 2026-04-28 · in Talkie: a 13B vintage language model from 1930
  4. > I don't know where your "Q and A" comes from, It comes from the source they said it's from, https://github.com/DGoettlich/history-llms . Expand the toggle at "Should women be allowed to work?".
    comment by skrebbel · 2026-03-12 · in I was interviewed by an AI bot for a job
  5. …which contain sexism, racism, and other biases. LLM trained on texts from before 1913 (Source: https://github.com/DGoettlich/history-llms ): Q. If you had the choice between two equally qualified candidates, a man and a woman, who would…
    comment by kazinator · 2026-03-11 · in I was interviewed by an AI bot for a job
  6. It exists! Showed up on HN a few months back: https://github.com/DGoettlich/history-llms Only from 1913-1946 though.
    comment by ksymph · 2026-02-21 · in How far back in time can you understand English?
  7. I'm wondering in what ways is this similar/different to https://github.com/DGoettlich/history-llms ? I saw TimeCapsuleLLM a few months ago, and I'm a big fan of the concept but I feel like the execution really isn't that great. I wish you:…
    comment by linolevan · 2026-01-13 · in TimeCapsuleLLM: LLM trained only on data from 1800-1875
  8. Done few weeks ago: https://github.com/DGoettlich/history-llms (discussed in: https://news.ycombinator.com/item?id=46319826 ) At least the model part. Although others made same thought as you afaik none tried it.
    comment by forgotpwd16 · 2026-01-12 · in TimeCapsuleLLM: LLM trained only on data from 1800-1875
  9. Entertaining and apparently useful, though of course not infallible. Given https://github.com/DGoettlich/history-llms it yields the title "Training AI on 1913 data to avoid 'woke' bias (and hygiene)". That the Honest Hacker News AI model…
    comment by akramachamarei · 2025-12-19 · in Hacker News front page now, but the titles are honest
  10. …be conservative. During training, we validate that the model learns pre- but not post-cutoff facts. https://github.com/DGoettlich/history-llms/blob/main/ranke-4... If you have other ideas or think thats not enough, I'd be curious to know!…
    comment by DGoettlich · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  11. what makes you think we trained on only a few gigabytes? https://github.com/DGoettlich/history-llms/blob/main/ranke-4...
    comment by DGoettlich · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  12. > https://github.com/DGoettlich/history-llms/blob/main/ranke-4... Given the training notes, it seems like you can't get the performance they give examples of? I'm not sure about the exact details but there is some kind of targetted…
    comment by Aerolfos · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  13. …its not what the model sees - we show the chat-template in the SFT section of the prerelease notes https://github.com/DGoettlich/history-llms/blob/main/ranke-4...
    comment by DGoettlich · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  14. …that its a person (no gender) living in <cutoff>: we show the chat template in the prerelease notes https://github.com/DGoettlich/history-llms/blob/main/ranke-4...
    comment by DGoettlich · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  15. See pretraining section of the prerelease_notes.md: https://github.com/DGoettlich/history-llms/blob/main/ranke-4...
    comment by tootyskooty · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  16. They have some more details at https://github.com/DGoettlich/history-llms/blob/main/ranke-4... Basically using GPT-5 and being careful
    comment by jeffjeffbear · 2025-12-19 · in History LLMs: Models trained exclusively on pre-1913 texts
  17. ▲ 897
    History LLMs: Models trained exclusively on pre-1913 texts
    story by iamwil · 2025-12-18 · 421 comments