trying different LLM

quickstart.py

@@ -3,16 +3,13 @@ from trl import GRPOTrainer
 import os
 
 dataset = load_dataset("trl-lib/tldr", split="train")
-assert os.path.isfile("trl-lib/tldr")
-with open("trl-lib/tldr", "r") as f:
-    pass
 
 # Dummy reward function: count the number of unique characters in the completions
 def reward_num_unique_chars(completions, **kwargs):
     return [len(set(c)) for c in completions]
 
 trainer = GRPOTrainer(
-    model="Qwen/Qwen2-0.5B-Instruct",
+    model="openai-community/gpt2",
     reward_funcs=reward_num_unique_chars,
     train_dataset=dataset,
 )