@@ -85,6 +85,7 @@ output_dir: ./completed-model
# training hyperparameters
batch_size: 8
micro_batch_size: 2
eval_batch_size: 2
num_epochs: 3
warmup_steps: 100
learning_rate: 0.00003
The note is not visible to the blocked user.