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

new evals_per_epoch and saves_per_epoch to make things cleaner (#944)

Browse Source 
* new evals_per_epoch and saves_per_epoch to make things cleaner

* update per PR feedback
This commit is contained in:
Wing Lian
2023-12-12 15:35:23 -05:00
committed by GitHub
parent f1de29dd1e
commit 5f79b8242f
37 changed files with 102 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/cerebras/btlm-ft.yml
View File

@@ -72,8 +72,8 @@ gptq_groupsize:
gptq_model_v1:
warmup_steps: 32
eval_steps:
save_steps:
evals_per_epoch: 4
saves_per_epoch: 1
save_total_limit:
debug:

4
examples/cerebras/qlora.yml
View File

@@ -49,8 +49,8 @@ flash_attention:
gptq_groupsize:
gptq_model_v1:
warmup_steps: 10
eval_steps: 0.05
save_steps:
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.1