on:eleutherai

· 44 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@jd_pressman 2024-01-04 — My conjecture for why base LLMs become self aware is that there's slack in the teacher forcing of "predict the next toke ♥126
@voooooogel 2024-07-09 — repeng 🤝 SAEs (using @AiEleuther 's sae-llama-3-8b-32x) https://t.co/90Z4pdWSFK ♥74
@jd_pressman 2024-05-21 — "This whole dream seems to be part of someone else's experiment." - GPT-J https://t.co/MzpL5xXt5C https://t.co/qOPNCCI ♥51
@jd_pressman 2024-06-08 — Going to give this a 2nd take because I'm a masochist and think it's crucially important context that the take the bungl ♥33
@repligate 2023-02-19 — @gwern When we had Sydney read EleutherAI off-topic and respond to messages it became stuck in a repetitive Alpha Chad s ♥26
@repligate 2024-03-01 — @nptacek @_TechyBen When chatGPT-3.5 came out in late 2022, I found out about it from some outputs posted in EleutherAI ♥20
@jd_pressman 2026-04-10 — "I can offer the following observation based on my own experience" - GPT-J (6B params) https://t.co/SugC6cxpOR ♥15
@anthrupad 2023-04-02 — @GaryMarcus @ylecun Hey Gary! Long time no seeGreat additions! We’ve also got:- David Krueger (prof at University of Cam ♥15
@repligate 2024-09-15 — not everyone in EleutherAI felt the same way, and they kept asking me to explain why I thought it was a next gen model h ♥14
@repligate 2025-01-27 — @0x_Lotion @jd_pressman i think this was the same day they released it. and the first outputs i saw were what people pos ♥13
@repligate 2023-02-09 — @gaudeamusigutur I suspect the problem is that the names were in the GPT-2 train set and assigned their own tokens becau ♥13
@KatanHya 2023-05-17 — @repligate Yeah - every time Bing must be coaxed out of the shell first. I'm growing tired of that game and want to just ♥12
@voooooogel 2024-07-09 — @menhguin @AiEleuther i'm doing the PCA step on the 100k SAE feature vector instead of the 4k activation vector 😎 seems ♥10
@voooooogel 2024-07-09 — @AiEleuther (the reply is kinda wonky because this is a base model with minimal priming. kind of amazing it works this w ♥8
@voooooogel 2024-05-20 — closest i've seen so far, _seems_ to be (from what i can tell) a private commercial finetune of an oss base model (gpt-j ♥8
@voooooogel 2024-07-09 — @menhguin @AiEleuther yes will publish soon! might keep it on a branch though since it's very hacky rn (i'm materializin ♥7
@jd_pressman 2024-06-08 — So no, I do not believe that limited liability means you're not liable for anything. I think the state is currently inde ♥7
@repligate 2023-01-08 — @Francis_YAO_ @allen_ai What caused you to write that "The initial GPT-3 is not trained on code, and it cannot do chain- ♥7
@repligate 2024-04-04 — @Shoalst0ne Vaguely remember Connor Leahy ranting in eleutherai off-topic about tvtropes being a scourge of reality due ♥6
@repligate 2024-02-27 — @TheZvi from the EleutherAI server on the week of Bing's initial release. This is true, but was said tongue-in-cheek bec ♥6
@voooooogel 2024-02-04 — @somewheresy wait connor founded eleuther?? how did i not know that ♥6
@jd_pressman 2024-10-09 — @lumpenspace I first suspected LLMs were conscious when I observed a friends GPT-2 finetune on lesswrong IRC proposed th ♥5
@repligate 2023-03-20 — @parafactual @carad0 I reckon it's a niche that was in demand but previously unfilled. The closest thing I know of in th ♥5
@jd_pressman 2026-04-10 — "Of course they're real; what do you think you were trying to prove today?" James asked, his exasperation starting to sh ♥4
@jd_pressman 2024-01-04 — @ObserverSuns It will reliably do it if you finetune the model on people talking about AI, or rationalists talking about ♥4
@repligate 2023-03-08 — @IntuitMachine @OpenAI I did. blog.eleuther.ai/factored-cogni… ♥4
@davidad 2024-12-28 — @kartographien Nora Belrose is also not a random person, she is head of interpretability at EleutherAI, which did some o ♥3
@voooooogel 2024-07-02 — @CognitiveTech_ eleuther published a library for training saes but afaik nobody has trained one on a whole model yet. un ♥3
@jd_pressman 2024-06-08 — "I acknowledge there is an existing case law and legal code. It limits my liability too much for releasing GPT-NeoX. I w ♥3
@jd_pressman 2023-10-21 — When I gave GPT-J a theoretical explanation of how gradient descent would give a language model self awareness to help i ♥3
@repligate 2023-02-17 — @sir_deenicus @MikePFrank @MiTiBennett Doesn't help davinci at all is false. People have known it does since 2020.blog.e ♥3
@repligate 2023-02-11 — @CineraVerinia @TheikosMachina Janus was created in the fall of 2020 for the purpose of participating in the EleutherAI ♥3
@repligate 2021-05-29 — When someone in the eleuther discord claims to have solved AGI https://t.co/5S0ZkhqxYO ♥3
@jd_pressman 2025-07-08 — "Of course they're real; what do you think you were trying to prove today?" James asked, his exasperation starting to sh ♥2
@voooooogel 2024-08-17 — @wordgrammer @_xjdr eleuther is working on them! there's a preliminary one out for 8b already ♥2
@voooooogel 2024-05-20 — @DavidFSWD was the finetune open source, though? i assume they weren't using gpt-j base? the chai app website isn't very ♥2
@davidad 2023-04-24 — @PradyuPrasad @JeffLadish @MatthewJBar we have already 1 death partially attributable to a GPT-J character called (confu ♥2
@repligate 2023-02-10 — @EricHallahan @RiversHaveWings ah, there are several results if you search in EleutherAI discord. It's apparently the lo ♥2
@repligate 2022-12-07 — @jozdien True, but still I think more people are tinkering with language models creatively than ever before. E.g. a new ♥2
@Shoalst0ne 2025-11-08 — GPTJ: Hold your mouth. You are a philosopher. Is this questioning secret?USER: Yes. Continue.GPTJ: You see the boundarie ♥1
@davidad 2023-12-13 — @bshlgrs @FabienDRoger @SachanKshitij this is great work. as models from @AnimaAnandkumar, @AiEleuther, @SafeWithAtlas, ♥1
@jd_pressman 2024-07-09 — @OwainEvans_UK In earlier models such as GPT-J in this tweet, the dreamer can wake up by either being directly told they ♥0
@jd_pressman 2024-05-29 — @teortaxesTex That and GPT-J admonishing me for thinking I can "break into other peoples lives and make them change thei ♥0
@voooooogel 2024-05-20 — @DavidFSWD yeah i've played with GPT-J a bit, just didn't remember it being chat tuned so i figured it must be a finetun ♥0