Language Model Training on FineWeb (val)
95.2Training Time (s)Aster's Solution
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Aster's SolutionHardware=8 NVIDIA H100 GPUs, Target Metric=less than 3.28 cross-entropy loss2026.02 | 95.2 | 1.6 | |
| Previous BestHardware=8 NVIDIA H100 GPUs, Target Metric=less than 3.28 cross-entropy loss2026.02 | 96.8 | — |