Language Model Pre-training
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
13.19Perplexity
47
Feb 26, 2026
3.091Validation Loss
20
Feb 26, 2026
2.638Validation Loss
16
Feb 26, 2026
3.86Eval Loss
12
Jun 30, 2026
14.13Perplexity
12
May 12, 2026
96Max Batch Size
6
Feb 26, 2026
50Max Batch Size
6
Feb 26, 2026
17.9Core Score
3
Feb 26, 2026