Pro@programming.dev to Programming@programming.devEnglish · elva dagar sedanSurprisingly Fast AI-Generated Kernels We Didn’t Mean to Publish (Yet)crfm.stanford.eduexternal-linkmessage-square5fedilinkarrow-up120arrow-down16
arrow-up114arrow-down1external-linkSurprisingly Fast AI-Generated Kernels We Didn’t Mean to Publish (Yet)crfm.stanford.eduPro@programming.dev to Programming@programming.devEnglish · elva dagar sedanmessage-square5fedilink
minus-squareSpicyToaster420@sopuli.xyzlinkfedilinkarrow-up4·tio dagar sedanAwesome use of LLMs. I wonder they didn’t use FP8 quantization though, especially since their target hardware was an L40s.
Awesome use of LLMs. I wonder they didn’t use FP8 quantization though, especially since their target hardware was an L40s.