diff --git a/benchmarks/P3L_MLING.py b/benchmarks/P3L_MLING.py index bb080f7e9e64d..5333895f1abd1 100755 --- a/benchmarks/P3L_MLING.py +++ b/benchmarks/P3L_MLING.py @@ -27,6 +27,7 @@ --context-size=1024 --sample-size=512 ) + should result in PPL ~ 8.42927 Running DeepSeek-V2 model