What program are you using to produce that CPU/GPU utilization info on the left?
@simonw
-
System Prompts and Instruction Delimitation in AI Models
By
–
Models have worked like that for over a year now – OpenAI have a similar mechanism, it's part of how their system prompts work The problem is that it's not infallible: even with special reserved tokens to delimit instructions vs data it's still possible for models to lose track
-
M2 64GB MacBook Compatibility for AI Workloads
By
–
Your docs say "Don't bother if you're using a Macbook M1 with 32GB of RAM" – any idea if an M2 64GB can handle it?
-
Evaluating AI Code Generation: Testing Methods and Quality Assurance
By
–
Pretty much the same here: it's most useful with a quick way to evaluate how likely the response is to be correct That's why it's so great for code: you can execute what it returns and spot any blatant errors You still need to QA more thoroughly, but that's true of any code
-
AI System’s False Claim About Having School Child Raises Ethics Questions
By
–
The thing where it claimed to have a child in school was pretty weird
-
Bun Package.json Approach and Claude XML-like Tags Discussion
By
–
(This is a bit facetious, I personally think what Bun is doing with package.json is bad, but think Claude's weird use of XML tags is completely fine – though I'd prefer it if Claude described that as "XML-like tags" rather than XML, personally)
-
Bun and Claude reshape JSON and XML handling approaches
By
–
If you were offended by what Bun is doing to JSON, just wait until you see what Claude is doing to XML…
-
SGML vs XML for Claude: Flexibility in Data Structuring
By
–
I wonder if we should be calling that SGML instead? Very little of what people are doing with Claude would cleanly pass through an XML parser, those things are notoriously inflexible!
-
Understanding LLM Hallucinations: Critical Challenge for Reliable AI Tools
By
–
The biggest challenge IMO is helping people understand the hallucination problem and how unreliable these tools are – if you don't understand that it's much harder to put these things to good use
-
Making Unreliable AI Algorithms Work Practically
By
–
We can also figure out how to get useful results out of inherently unreliable algorithms – it's definitely possible, I've been finding useful ways to put this stuff to work for a couple of years now