tangledgroup
/

tangled-alpha-0.9-core

Text Generation

Model card Files Files and versions Community

mtasic85 commited on Mar 15

Commit

ebd7609

·

1 Parent(s): a2e4903

cpt core 4

Files changed (1) hide show

scripts/cpt_core_model_4.py +1 -1

scripts/cpt_core_model_4.py CHANGED Viewed

@@ -96,7 +96,7 @@ trainer = UnslothTrainer(
         # gradient_accumulation_steps=64,
         # per_device_train_batch_size=16,
         # gradient_accumulation_steps=16,
-        per_device_train_batch_size=2,
         gradient_accumulation_steps=8,
         warmup_ratio=0,

         # gradient_accumulation_steps=64,
         # per_device_train_batch_size=16,
         # gradient_accumulation_steps=16,
+        per_device_train_batch_size=1,
         gradient_accumulation_steps=8,
         warmup_ratio=0,