cgorlla
Joined 6 years ago
Karma
153
Submissions
42
Member since
Apr 29
Recent submissions
Comment · 2 months ago
Very likely, NYT confirmed it's releasing Friday.
Comment · 2 months ago
It's decent. How good depends on compute needs.
Comment · 2 months ago
It's fun! Also it's an interesting commentary on where the AI industry is as a whole.
Comment · 2 months ago
Based on the reception of our last post we took folks' suggestions to run the new official build of V4 Flash and compare it to the preview that was released 5 days apart. The new build is added to https://playground.ctgt.ai/ and you can run one comparison with no email verification if you're curious.
We ran LineageEval with the same setup and found the official was 6.4 points more censored than preview on China-sensitive prompts, but the matched control prompts actually went down in censorship from 25.4 to 19.8.The matched gap widened 12 points, from +32.0 to +44.0. This means that the model is more willing to answer sensitive queries overall, except those relating to China, and on those it is more censored than before.
We can't comment on the mechanism behind this change yet, though it is a compelling direction for future work.
We reran the distillation run on the finance objective with 2 new teacher models, Inkling Small which was the least censored model we've tested, and V4-0731 which was the most. The teachers spanned a 5.5x range but the students all remained similar to their base models.
Thanks to a commenter from last time for flagging SpeechMap.ai. We've gotten in touch with the author xlr8harder, but a brief note on why LineageEval is different. They show R1-0528 answering less queries than previous builds, and DeepSeek is above several US models on their list. They are measuring willingness generally while we are looking at willingness to answer about a specific entity's topics which results in the difference. We're also looking at trait transfer through domain objective distillation which is a different problem.
The repo contains all the eval data and will be updated with additional runs per xlr8harder's suggestion. https://github.com/CTGT-Inc/lineage-eval
Comment · 2 months ago
Every post is actually scanned for LLM content as well.
Comment · 2 months ago
A typographic Voight-Kampff test is pretty awesome.
Comment · 3 months ago
We discuss this in the writeup. While we expected this result, it is important for there to be data backing the claims, and an experimental setup that mirrors productions tasks is a useful tool for the conversations going on about this.
Comment · 3 months ago
We're actually exploring the changes in the model geometry that cause it to comply or not comply with a given policy next, I think visual representations of that behavior would be interesting and perhaps elucidating. What you mention is also a worthy line of work.
Comment · 3 months ago
I guess the AI that wrote your comment for you also conflated the SFT step of the target domain with the political prompts, which, in the sentence you quoted, contradicts your original comment...
Comment · 3 months ago
>We plan to test what happens with a Chinese teacher into a Chinese-lineage base like Qwen next.
:)
Comment · 3 months ago
The fact that you literally thought the examples were used in SFT in your last comment ago calls into question the utility of this conversation, notwithstanding the implication that those examples were used to improve…financial performance?
This is a very standard setup for a distillation problem. The vast majority of companies don't care about the "wide concept", this is what most distillation consists of. They want to improve models on a narrow domain. It should be understood that this is by and large a low risk vector for this sort of behavior to transfer. That is what we are measuring, and we are very open about it.