address PR feedback

2023-06-10 14:21:43 -04:00
parent eea2731a5e
commit 0c6f928601
5 changed files with 9 additions and 8 deletions
--- a/scripts/finetune.py
+++ b/scripts/finetune.py
@@ -208,7 +208,10 @@ def train(
            )
        else:
            train_dataset = load_pretraining_dataset(
-                cfg.pretraining_dataset, tokenizer, max_tokens=cfg.sequence_len
+                cfg.pretraining_dataset,
+                tokenizer,
+                max_tokens=cfg.sequence_len,
+                seed=cfg.seed,
            )
            # https://discuss.huggingface.co/t/how-to-use-huggingface-trainer-streaming-datasets-without-wrapping-it-with-torchdatas-iterablewrapper/25230
            train_dataset = train_dataset.with_format("torch")