bug fix

Author: chenyushuo

trinity/trainer/tinker_trainer.py

@@ -90,7 +90,7 @@ async def prepare(self):
                 self._train_step_num = self.latest_remote_checkpoint_step = int(f.read().strip())
             checkpoint_file_path = os.path.join(
                 self.default_local_dir,
-                f"global_step_{self.train_step_num}",
+                f"global_step_{self._train_step_num}",
                 "remote_checkpoint_path.txt",
             )
             with open(checkpoint_file_path, "r") as f:

trinity/trainer/trainer.py

@@ -59,7 +59,7 @@ def __init__(self, config: Config) -> None:
             sample_strategy_state = trainer_state.get("sample_strategy_state", {})
         self.sample_strategy.load_state_dict(sample_strategy_state)
         self.save_interval = config.trainer.save_interval
-        self.last_sync_step = self.train_step_num
+        self.last_sync_step = 0
         self.last_sync_time = None
         self.total_steps = config.trainer.total_steps or float("inf")
         self.save_hf_checkpoint = config.trainer.save_hf_checkpoint