Language Modeling on Upstream
0.0715CFCF
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CFModel=Transformer, Scaling=Constant2026.05 | 0.0715 | — | — | — | |
| CFModel=Recurrent, Scaling=Constant2026.05 | 0.0313 | — | — | — | |
| CFModel=Transformer, Scaling=Chinchilla2026.05 | 0.0172 | — | — | — | |
| A1Model=Transformer, Scaling=Chinchilla2026.05 | — | 0.0099 | — | — | |
| A1Model=Transformer, Scaling=Constant2026.05 | — | 0.0398 | — | — | |
| A1Model=Recurrent, Scaling=Constant2026.05 | — | 0.0307 | — | — | |
| A2Model=Transformer, Scaling=Chinchilla2026.05 | — | — | 0.0044 | — | |
| A2Model=Transformer, Scaling=Constant2026.05 | — | — | 0.0289 | — | |
| A2Model=Recurrent, Scaling=Constant2026.05 | — | — | 0.0192 | — | |
| UNSLModel=Transformer, Scaling=Chinchilla2026.05 | — | — | — | 0.0038 | |
| UNSLModel=Transformer, Scaling=Constant2026.05 | — | — | — | 0.008 | |
| UNSLModel=Recurrent, Scaling=Constant2026.05 | — | — | — | 0.0047 |