LLM Terminology explained simply: Weights, Inference, Effective sequence lengthdevforth.io5 ポイント·投稿者 dotnot·4 か月前·1 コメント
Real-world GPT-OSS-20B benchmark on L4, L40S and H100 (latency, tokens/SEC)devforth.io1 ポイント·投稿者 dotnot·5 か月前·0 コメント
dotnot·4 か月前·議論Nice readme example, I would also like to see time of execution for these tasks, I mean how much time for handling filters adds comparing to non using it