tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

fix: set replit mpt model to use eager attention

Browse Source
This commit is contained in:
NanoCode012
2025-05-07 16:55:01 +07:00
parent 6ee7cb30fa
commit d0c4930dd5
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/replit-3b/config-lora.yml
View File

@@ -38,7 +38,7 @@ tf32: true
gradient_checkpointing: gradient_checkpointing:
resume_from_checkpoint: resume_from_checkpoint:
logging_steps: 1 logging_steps: 1
attention: flash attention: eager
gptq_groupsize: gptq_groupsize:
gptq_model_v1: gptq_model_v1:
warmup_steps: 20 warmup_steps: 20