buppermint

7 ms·

1.
by buppermint 13d ago
CoT is both correlated and causal of the model's real computations, it's just imperfect. If you manually add "Let's wrap it up" in the CoT during generation, most LLMs will actually wrap it up (this is a commonly
2.
by buppermint 4mo ago
The paper title is a bit misleading. The tested detectors and models here are small and rather dated (Llama 3.1 8B and Gemini Flash 2.0 - these are basically in the level of a modern 1B model), and the actual paper says this only shows vuln
3.
by buppermint 5mo ago
Worth seeing the whole chart in perspective: https://fred.stlouisfed.org/series/IHLIDXUSTPSOFTDEVE
4.
by buppermint 8mo ago
Anthropic has already has lower guardrails for DoD usage: https://www.theverge.com/ai-artificial-intelligence/680465/a... It's interesting to me that a company that claims to be all about the public good: - S
5.
by buppermint 8mo ago
I would disagree on the knowledge sharing. They're the only major AI company that's released zero open weight models. Nor do they share any research regarding safety training, even though that's supposedly the whole reason fo
6.
by buppermint 9mo ago
I've been playing around with this in z-ai and I'm very impressed. For my math/research heavy applications it is up there with GPT-5.2 thinking and Gemini 3 Pro. And its well ahead of K2 thinking and Opus 4.5.
7.
by buppermint 9mo ago
From a quick read, this is cool but maybe a little overstated. From Figure 3, completely suppressing these neurons only reduces hallucinations by like ~5% compared to their normal state. Table 1 is even more odd, H-neurons predicts hallucin
8.
by buppermint 9mo ago
This is because Grok Code Fast is free via Kilo Code/Cline and has been for months
9.
by buppermint 11mo ago
Very cool. Any plans to add support for local models? This has what has prevented us from adopting Positron so far. We have sensitive data and sending to third party APIs is not an option (regardless of their stated retention policies).