From 648af26073058df66a98b58f98e37e2e4d260974 Mon Sep 17 00:00:00 2001 From: Eric Wang Date: Tue, 14 Mar 2023 08:51:30 -0700 Subject: [PATCH] update hyperparams --- finetune.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/finetune.py b/finetune.py index 10876d68..673b3f95 100644 --- a/finetune.py +++ b/finetune.py @@ -62,8 +62,8 @@ def generate_prompt(data_point): MICRO_BATCH_SIZE = 4 # this could actually be 5 but i like powers of 2 BATCH_SIZE = 128 GRADIENT_ACCUMULATION_STEPS = BATCH_SIZE // MICRO_BATCH_SIZE -EPOCHS = 3 # from the result -LEARNING_RATE = 2e-5 # also from the result +EPOCHS = 1 # we don't need 3 tbh +LEARNING_RATE = 3e-4 # the karpathy constant CUTOFF_LEN = 256 # 256 accounts for about 96% of the data data = data.shuffle().map(