200,000 Subscribers
Just saw that my blog reached 200,000 subscribers.
A big thank you to everyone who has been reading over the years and for all the encouraging feedback on my posts. It really kept me motivated to continue writing these long-form articles.
And an extra thank you to those who support this journey with a paid subscription. I really appreciate it. Your support makes this work as an independent researcher possible.
P.S. I am currently working on an exciting new article. If things go smoothly, it should be out next weekend.
Source: lightly edited website version of a note from my blog.
Read Next
Focusing on Post-Training
Why I'd invest in post-training existing open-weight LLMs, with Fireworks' Ember-1 as an example of more token-efficient reasoning.
MiMo-V2.6 Pro Architecture and Training Notes
Notes on MiMo-V2.6 Pro's GQA and sliding-window attention, agent training tasks, reward signals, and large RL batches.
It's Easy to Dismiss Jev as Just a Classifier
A short note on Jev's generalization, possible encoder-style architecture and training, and Choice and Noul API examples.