← Back to Test-Time Training
score-first SGD
Test-Time TrainingUsed in
1 PRs
Best BPB
1.0810
Avg BPB
1.0810
Submissions
Hyperparameters Across PRs
| pr_number | parameters |
|---|---|
| 1658 | {"epochs_per_chunk":3,"chunk_size":32000,"learning_rate":0.005,"momentum":0.9,"grad_clip":1} |