chococoan
/

llm-jp-3-13b-it_lora3

text-generation-inference

Model card Files Files and versions Community

chococoan commited on Dec 16, 2024

Commit

9419630

·

verified ·

1 Parent(s): 8c19325

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -106,11 +106,11 @@ trainer = SFTTrainer(
     dataset_text_field="formatted_text",
     packing = False,
     args = TrainingArguments(
-        per_device_train_batch_size = 2,
         gradient_accumulation_steps = 4,
         num_train_epochs = 1,
         logging_steps = 10,
-        warmup_steps = 10,
         save_steps=100,
         save_total_limit=2,
         max_steps=-1,

     dataset_text_field="formatted_text",
     packing = False,
     args = TrainingArguments(
+        per_device_train_batch_size = 8,
         gradient_accumulation_steps = 4,
         num_train_epochs = 1,
         logging_steps = 10,
+        warmup_steps = 100,
         save_steps=100,
         save_total_limit=2,
         max_steps=-1,