AI

For me, it was seeing some of the creative stuff people were making with image generators that really blew my mind. I'm curious what sparked everyone else's interest and got them exploring this field.

A recent paper from Anthropic highlights an emergent internal 'workspace' within language models. These models generate silent words they can utilize for reporting, steering, and reasoning. A fascinating example is when a model is asked to evaluate '12 + 5 = 1'. It internally…

Honestly, I think it's the sheer amount of manual data labeling and curation that still happens behind the scenes. Everyone's obsessed with the flashy models, but the quality of the data going in is what actually makes or breaks them. Am I the only one who feels like this dirty…

I used to think that fine tuning was the only way to make a model useful for specific tasks, but now I'm all in on clever prompting and retrieval augmented generation. It's wild how much you can get out of a general model with the right context and a well structured query. Has…

I keep finding myself pulling up the same few papers and blog posts whenever I need to refresh my understanding of transformer architectures. What about you: what’s your go to when you need a reliable refresher or a new perspective?