Language Modeling on StoryCloze
57.56Accuracy (%)GPT-2 small
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-2 smallARACH=true, Logit offset (b)=-0.42026.03 | 57.56 | |
| GPT-2 smallARACH=true, Logit offset (b)=-0.52026.03 | 57.56 | |
| GPT-2 smallARACH=true, Logit offset (b)=-0.32026.03 | 57.51 | |
| GPT-2 smallARACH=false2026.03 | 57.46 |