then ship a lora i understand the confusion but have thought thru this one and i'm p sure i'm correct – will just do a longer writeup. coming soon (this week)
LLMS
-
Running Evals in Headless Chrome for Gemini Nano
By
–
@grmcameron could @ArtificialAnlys run evals in headless chrome for gemini nano?
-
Preference for terminology in AI prompt and context engineering
By
–
I've never loved "prompt engineering" but I don't really like "context engineering" either. You?
-
Reduce o3 Hallucinations: Citation Prompt Engineering Strategy
By
–
Add this to your prompt to reduce o3 hallucinations: "Cite any information included in your response inline, referencing the source where you originally found it. Include a snippet of the original text, verbatim, so I can verify it."
-
LLM Evaluation for Agentic Systems: Chris Borg’s Talk
By
–
Couldn’t make it to @databricks #DataAISummit? Chris Borg’s talk on LLM evaluation for agentic systems is live. Full video: https://
youtube.com/watch?v=tlEmxr
qKGlE
… #SnorkelAI #GenAI #LLMEval -
ChatGPT messes up your brain intentionally, video analysis
By
–
Comment #ChatGPT is messing up your brain? Full analysis → https://
youtu.be/mZDhRjzV1Hk The worst… It's that it's intentional -
Prompting Grok to Summarize Every Book in One Word
By
–
« OK Grok, résume chaque livre jamais écrit en un seul mot. »
-
SmolLM3-3B fills Qwen’s Pareto gap via agentic post-training
By
–
Qwen left a hole in the Pareto frontier of optimal performance for a given size… So we just filled it: introducing SmolLM3-3B I helped the SmolLM team on the "make it agentic" part, by post-training the model on agent traces with @akseljoonas
: the model is now also on -
Quantization Update: Demand Scaling or Silent Model Degradation?
By
–
Quantization to support more demand, or could also be a silent update gone wrong.
-
Production Use of Sonnet at Magicpath AI
By
–
I use Sonnet mostly in production at @Magicpathai and rarely use Opus, so I'm referring to Sonnet in particular.