← Back to Test-Time Training
MLP-down-only TTT
Test-Time TrainingUsed in
1 PRs
Best BPB
1.1142
Avg BPB
1.1142
Submissions
Hyperparameters Across PRs
| pr_number | parameters |
|---|---|
| 756 | {"learning_rate":0.002,"epochs":3,"chunk_tokens":32768,"stride":64} |