Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Language Modeling on UL2 Pre-training (val)
Loading...
1.841
Loss
Baseline
1.83284
1.88792
1.943
1.99808
Sep 27, 2024
Loss
Updated 4d ago
Evaluation Results
Method
Method
Links
Loss
Baseline
Parameters=8B, d(i)/i...
2024.09
1.841
MIDAS
Parameters=8B, d(i)/i...
2024.09
1.844
Baseline
Parameters=2B, d(i)/i...
2024.09
1.926
MIDAS
Parameters=2B, d(i)/i...
2024.09
1.929
GRADSTACK
Parameters=2B, d(i)/i...
2024.09
1.945
MIDAS
Parameters=2B, d(i)/i...
2024.09
1.947
Baseline
Parameters=1B, d(i)/i...
2024.09
1.996
MIDAS
Parameters=1B, d(i)/i...
2024.09
1.999
MIDAS
Parameters=1B, d(i)/i...
2024.09
2.009
MIDAS
Parameters=1B, d(i)/i...
2024.09
2.012
GRADSTACK
Parameters=1B, d(i)/i...
2024.09
2.024
MIDAS
Parameters=1B, d(i)/i...
2024.09
2.028
MIDAS
Parameters=1B, d(i)/i...
2024.09
2.032
GRADSTACK
Parameters=1B, d(i)/i...
2024.09
2.045
Feedback
Search any
task
Search any
task