← Back to Architecture
larger MLP
ArchitectureUsed in
1 PRs
Best BPB
1.2236
Avg BPB
1.2236
Submissions
Hyperparameters Across PRs
| pr_number | parameters |
|---|---|
| 812 | {"mlp_multiplier":2.65} |
| pr_number | parameters |
|---|---|
| 812 | {"mlp_multiplier":2.65} |