OpenMOSE commited on
Commit
65cea44
·
1 Parent(s): 646f063

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -2
README.md CHANGED
@@ -71,8 +71,9 @@ curl http://127.0.0.1:9000/loadmodel -X POST -H "Content-Type: application/json"
71
  ## Training Infrastructure
72
  - Hardware: Single AMD MI300X GPU
73
  - Training Duration: 3 days(Stage1,2)
74
- - Stage1 180MToken
75
- - Stage2 160MToken (in training)
 
76
  - Stage3 1GToken(TBD)
77
 
78
  ## Acknowledgements
 
71
  ## Training Infrastructure
72
  - Hardware: Single AMD MI300X GPU
73
  - Training Duration: 3 days(Stage1,2)
74
+ - Stage1 180MToken (LR1e-4)
75
+ - Stage2 160MToken (Temperature 1.0KD LR5e-6)
76
+ - Stage2.5 TBD (Temperature 2.0KD LR3e-5)
77
  - Stage3 1GToken(TBD)
78
 
79
  ## Acknowledgements