← Back to Architecture
GPT depth increase
ArchitectureUsed in
1 PRs
Best BPB
1.1407
Avg BPB
1.1407
Submissions
Hyperparameters Across PRs
| pr_number | parameters |
|---|---|
| 845 | {"layers":12} |
| pr_number | parameters |
|---|---|
| 845 | {"layers":12} |