← Back to Architecture
memory tokens
ArchitectureUsed in
2 PRs
Best BPB
1.1466
Avg BPB
1.4994
Hyperparameters Across PRs
| pr_number | parameters |
|---|---|
| 345 | {"count":16} |
| 421 | {"tokens":64} |
| pr_number | parameters |
|---|---|
| 345 | {"count":16} |
| 421 | {"tokens":64} |