Add padding_token=eos_token for GPU trl QLora example (#12398)

* Avoid tokenizer doesn't have a padding token error.
intel-analytics · Nov 14, 2024 · 7e50ff1 · 7e50ff1
1 parent d2cbcb0
commit 7e50ff1
Showing 1 changed file with 3 additions and 0 deletions.
diff --git a/python/llm/example/GPU/LLM-Finetuning/QLoRA/trl-example/qlora_finetuning.py b/python/llm/example/GPU/LLM-Finetuning/QLoRA/trl-example/qlora_finetuning.py
@@ -45,6 +45,9 @@
     model_path = args.repo_id_or_model_path
     dataset_path = args.dataset
     tokenizer = LlamaTokenizer.from_pretrained(model_path, trust_remote_code=True)
+    # Avoid tokenizer doesn't have a padding token
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token
 
     if dataset_path.endswith(".json") or dataset_path.endswith(".jsonl"):
         data = load_dataset("json", data_files=dataset_path)