Mathematical Reasoning (Fine-tuning) on GSM8K (test)
4.598Test PerplexityBAOC0.5
Evaluation Results
| Method | Links | |
|---|---|---|
| BAOC0.5Architecture=GPT2-small, Time (min)=30.8, StateMem (MB)=2352026.05 | 4.598 | |
| AdamW16Architecture=GPT2-small, Time (min)=31.5, StateMem (MB)=4752026.05 | 4.615 | |
| COSMOSArchitecture=GPT2-small, Time (min)=40.3, StateMem (MB)=6602026.05 | 4.737 | |
| Adam-miniArchitecture=GPT2-small, Time (min)=30.2, StateMem (MB)=4752026.05 | 4.995 | |
| MuonArchitecture=GPT2-small, Time (min)=35.8, StateMem (MB)=6252026.05 | 6.238 | |
| Adam-miniArchitecture=T5-base, Time (min)=28.9, StateMem (MB)=7562026.05 | 8.898 | |
| BAOC0.5Architecture=T5-base, Time (min)=22.8, StateMem (MB)=3572026.05 | 9.137 | |
| AdamW16Architecture=T5-base, Time (min)=28.2, StateMem (MB)=7562026.05 | 9.312 | |
| COSMOSArchitecture=T5-base, Time (min)=61.3, StateMem (MB)=8582026.05 | 9.921 | |
| MuonArchitecture=T5-base, Time (min)=39.2, StateMem (MB)=9452026.05 | 13.049 |