beomi commited on
Commit
e938116
1 Parent(s): 77aa647

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -34,7 +34,7 @@ Llama-2-Ko is an auto-regressive language model that uses an optimized transform
34
 
35
  ||Training Data|Params|Content Length|GQA|Tokens|LR|
36
  |---|---|---|---|---|---|---|
37
- |Llama 2|*A new mix of Korean online data*|7B|4k|&#10007;|160B|3.0 x 10<sup>-4</sup>|
38
 
39
  **Vocab Expansion**
40
 
 
34
 
35
  ||Training Data|Params|Content Length|GQA|Tokens|LR|
36
  |---|---|---|---|---|---|---|
37
+ |Llama 2|*A new mix of Korean online data*|7B|4k|&#10007;|>20B|1e<sup>-5</sup>|
38
 
39
  **Vocab Expansion**
40