Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Sequence Modeling on Minipile (pretraining)
Loading...
17.32
Perplexity
LayUp
16.9452
19.4751
22.005
24.5349
Oct 8, 2024
Perplexity
Training Time (s)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Perplexity
Training Time (s)
LayUp
Architecture=GPT-2 Med...
2024.10
17.32
38,624.33
GoSGD
Architecture=GPT-2 Med...
2024.10
17.52
41,147.67
SlowMo
Architecture=GPT-2 Med...
2024.10
17.72
52,687
CO2
Architecture=GPT-2 Med...
2024.10
17.75
51,508
DDP
Architecture=GPT-2 Med...
2024.10
17.85
40,658.05
AD-PSGD
Architecture=GPT-2 Med...
2024.10
26.69
38,093.67
Feedback
Search any
task
Search any
task
