issues
search
mrseanryan
/
gpt-eval-translations
Evaluate translations by either a self-hosted Embedder or using Chat-GPT as LLM-as-judge.
MIT License
0
stars
0
forks
source link
Add eval of the multi-lingual vs the standard embedder
#2
Closed
mrseanryan
closed
3 months ago
mrseanryan
commented
3 months ago
use golden dataset - see #1
mix up programmatically
translate and gen csv
eval
mrseanryan
commented
3 months ago
negative test for same-same cases
mrseanryan
commented
3 months ago
done