tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

fix eval_steps to be a sane default (#797)

Browse Source 
* fix eval_steps to be a sane default

* update docs for fractional eval_steps
This commit is contained in:
Wing Lian
2023-10-27 22:36:30 -04:00
committed by GitHub
parent 0800885e2f
commit 8b79ff0e94
23 changed files with 36 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/jeopardy-bot/config.yml
View File

@@ -24,7 +24,7 @@ wandb_log_model:
output_dir: ./jeopardy-bot-7b
gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 3
num_epochs: 4
optimizer: adamw_bnb_8bit
torchdistx_path:
lr_scheduler: cosine