Musk Viewer
About
Privacy Policy
Removal Request
Matthew Leavitt
@leavittron
1 year
Aran Komatsuzaki
@arankomatsuzaki
1 year
To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis Studies what would happen if we train LLM with repeated data and how we can alleviate the LLM mult-epoch degradation.
2
19
128
Download
0
0
39
Download