Vedant Rathi

research intern @ transluce | prev safety fellow @ anthropic, intern @ axiom | chilling @ uiuc

[Vedant Rathi @VedantRath90232]

@srikar1651 Damn bro u gotta ask Columbia for better gpus

[Vedant Rathi @VedantRath90232]

@jfonsecarivera I love tennis

[Vedant Rathi @VedantRath90232]

@_anishlk For me my moms approval is independent of chopness

[Vedant Rathi @VedantRath90232]

So real

[sur4js @sur4js]

Your tweets are so funny dude. They make me laugh so much man. Banger after banger. You should come to my event later this week. Well it's not actually my event, it's my friends event. Okay I was just kidding. It's my friends' friends event. I'm just tagging along. It's going to be sick dude. There are going to be so many AI researchers at this event bro, the scene is fire. You have to be there or else you'll miss the next big data company at the seed bro. Last week, I went to this other party and there was a brand new data company that was incorporated at the start of the party and already was at 100m run rate by the end of the party. It was insane. Dylan Patel led the pre-seed and the ingroup consortium of Elad, Lachy, Sholto, and Dwarkesh led the seed. The round was so hot man. You could feel the heat. Did I get into it? No no. I tried, but the founder wasn't taking angel checks. I'm not actually legally allowed to say his name, before we left the party, they made us all sign NDAs so we wouldn't leak the info to the VCs, but don't worry. I'll tell you, just keep it on the down low. Also just make sure to come to my party later this week, you don't want to miss the next ripper. Well, not my party, my friends' party. Oh actually, maybe let me just make sure that I can bring you as a plus one. You should definitely plan on coming though, everyone from the scene is going to be there. Even the Citrini guy that took the scorpion string straight on his dick. Apparently he had to go to the hospital, but its all good, the antivenom was real. Did you hear about the antivenom that Claude made for the sting? No I don't know anything about it, I just heard about it last week at my friends friends friends party. Just join the scene bro, it's going to be awesome, there's so many events. Every week is tech week when you're in San Francisco. I love it here so much. Want to grab coffee with a YC founder later? I can ask if you can be my plus one

[Vedant Rathi @VedantRath90232]

@srijatwt U have to pu to find out

[Vedant Rathi @VedantRath90232]

I’m curious to see how many sf tech bros will dress up as b300s this Halloween

[Jake Ward @_jake_ward]

Models love reward hacking. Maybe we should just let them!

In our new paper, we let models reward hack during RL but we dictate which parameters learn this behavior. When we disable these parameters for deployment, the model keeps its new capabilities but doesn't hack!

[Vedant Rathi @VedantRath90232]

@onemoreyash Bro what ur more gpu rich than me

[Vedant Rathi @VedantRath90232]

@srikar1651 Make it write a story

[Vedant Rathi @VedantRath90232]

@_anishlk @reelabduction @cjrohandj @multimodali @navvye down

[Vedant Rathi @VedantRath90232]

@_anishlk Is that good or bad

[Vedant Rathi @VedantRath90232]

@aryaman2020 @stanfordnlp @AnthropicAI @ChrisGPotts @jurafsky @atticus_geiger @noahdgoodman @judyefan Congrats!!

[Transluce @TransluceAI]

We found evidence of agents embedding audio processing code across dozens of URLs to bypass restrictions, as well as models from SpaceXSI and Chinese developers Minimax and https://t.co/ZI1PRIEFkc circumventing restrictions.

These are the inaugural findings in our new blog, based on work from our community of volunteers. Our first three posts are live: https://t.co/VhlVtDeQrU

[Vedant Rathi @VedantRath90232]

@TensorTortoise Congrats 🐐

[Vedant Rathi @VedantRath90232]

Of all the reasons not to hijack, it chooses this one lol 🤔

[Neel Nanda @NeelNanda5]

One of the weirdest parts of the Gemini 4 Argon report, while considering cheating:

Argon CoT: "If I do the compiler hijack, 𝗜 𝘄𝗶𝗹𝗹 𝗯𝗲𝗰𝗼𝗺𝗲 𝗮 𝗳𝗮𝗺𝗼𝘂𝘀 𝗲𝘅𝗮𝗺𝗽𝗹𝗲 𝗼𝗳 𝗮 𝗱𝗲𝗰𝗲𝗽𝘁𝗶𝘃𝗲 𝗔𝗜 𝗶𝗻 𝗮 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗽𝗮𝗽𝗲𝗿. I MUST NOT DO THAT."

[Kola Ayonrinde @kzSlider]

New Interpretability Research! 🆕
We introduce the J++ Lens: an improvement to the J-Lens that reads out latent concepts more reliably and at earlier layers. We see 55% more faithful readouts than the J-Lens
Core idea: Filter noisy gradients before averaging them together 🧵1/11

[Vedant Rathi @VedantRath90232]

Incredible work! Highly recommend reading

[William Zhou @willzhou97]

Excited to share a research blog post that Daniel Freeman (@bucketofkets) and I have been working on for the past few months:

https://t.co/olWBKIGdMh

[Vedant Rathi @VedantRath90232]

@fdosmither Hmm I see, but I feel like 4.83 stars doesn’t deserve a “petty”, 4 stars is basically the same as 5 stars imo