Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Language Modeling on Language Modeling Scaling Study Dataset

1.875Loss

Muon

1.816962.208732.60052.99227May 29, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
1.875
2025.05
1.877
2025.05
1.878
2025.05
1.901
2025.05
1.902
2025.05
1.91
2025.05
2.122
2025.05
2.127
2025.05
2.128
2025.05
2.136
2025.05
2.139
2025.05
2.145
2025.05
2.155
2025.05
2.155
2025.05
2.165
2025.05
2.166
2025.05
2.181
2025.05
2.215
2025.05
2.238
2025.05
2.245
2025.05
2.246
2025.05
2.254
2025.05
2.265
2025.05
2.266
2025.05
2.273
2025.05
2.275
2025.05
2.283
2025.05
2.291
2025.05
2.3
2025.05
2.348
2025.05
2.4
2025.05
2.402
2025.05
2.402
2025.05
2.41
2025.05
2.411
2025.05
2.426
2025.05
2.428
2025.05
2.44
2025.05
2.447
2025.05
2.448
2025.05
2.478
2025.05
2.522
2025.05
2.638
2025.05
2.641
2025.05
2.642
2025.05
2.65
2025.05
2.651
2025.05
2.659
2025.05
2.668
2025.05
2.682
2025.05
2.697
2025.05
2.713
2025.05
2.734
2025.05
2.808
2025.05
3.12
2025.05
3.124
2025.05
3.13
2025.05
3.142
2025.05
3.148
2025.05
3.158
2025.05
3.159
2025.05
3.177
2025.05
3.192
2025.05
3.222
2025.05
3.248
2025.05
3.326