support non-reenterent-gc & fix #6358

2025-12-15 11:20:35 +08:00 · 2024-12-17 11:41:59 +00:00
parent 81815f053f
commit f319da6937
2 changed files with 7 additions and 1 deletions
--- a/src/llamafactory/hparams/model_args.py
+++ b/src/llamafactory/hparams/model_args.py
@@ -237,6 +237,10 @@ class ModelArguments(QuantizationArguments, ProcessorArguments, ExportArguments,
        default=False,
        metadata={"help": "Whether or not to disable gradient checkpointing."},
    )
+    use_reentrant_gc: bool = field(
+        default=True,
+        metadata={"help": "Whether or not to use reentrant gradient checkpointing."},
+    )
    upcast_layernorm: bool = field(
        default=False,
        metadata={"help": "Whether or not to upcast the layernorm weights in fp32."},