# Saving weights while finetuning is on

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsaving-weights-while-finetuning-is-on%2F92050%3E%5C
**Category:** DeepSpeed\
**Created:** [June 13, 2024, 6:08pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsaving-weights-while-finetuning-is-on%2F92050 "2024-06-13T18:08:14Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![arihant-neohumans](https://avatars.discourse-cdn.com/v4/letter/a/fbc32d/32.png) [@arihant-neohumans](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Farihant-neohumans)%5C
**Post date:** [June 13, 2024, 6:08pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsaving-weights-while-finetuning-is-on%2F92050%2F1 "2024-06-13T18:08:14Z")

</div>

I have started finetuning using SFT and deepspeed. I have an automatic saving strategy at 500 steps. However, my calculations have gone wrong, and trying the model will take time. Is there any way that I can access the last checkpoint of the model at any time during training?