← Back to Test-Time Training

full TTT with SGD

Test-Time Training
Used in
1 PRs
Best BPB
1.1207
Avg BPB
1.1207

Hyperparameters Across PRs

pr_numberparameters
577{"learning_rate":0.002,"epochs":3,"max_train_chunks":50,"EMA_decay":0,"freeze_blocks":2,"optimizer":"SGD"}