b12498
server: slot_sched: fix lora and recurrent ( #32434 ) server: fix lora and recurrent tests: run the hybrid SWA checkpoint test with one token per ubatch With and without the cache, the prompt was not split into the same ubatches, and the generated tokens are decoded one at a time while the uncached run processes them in a batch. On CUDA this moves the top logprobs of the tiny model by up to 0.03 a
正在把原文抓到本机服务器…首次约 3~8 秒,抓好后本站永久缓存,下次秒开。
抓不到时可直接看原文(部分站点需要登录/反爬,服务器也无法抓取)。
正文由本站服务器抓取保存(已自动过滤广告、推荐位与追踪图片),排版与原站可能不同;版权归原站所有。