Reviewing Post-Training Techniques: DeepSeek, Qwen-2, and Phi-4brianfitzgerald.xyz·2 pts·roborovskis·0
Tricks for DPO Tuning a Code LLM, Part 1: Logit Curriculum Learningbrianfitzgerald.xyz·2 pts·roborovskis·0
Implementation of a Diffusion Transformer and Rectified Flow Sampling in Jaxgithub.com·1 pts·roborovskis·0
SuperPrompt: Better Text to Image Prompts in 77M Parametersbrianfitzgerald.xyz·150 pts·roborovskis·31