shrink-v1 / README.md
crumb's picture
+metrics: arc, tqa +metadata
bbe24b3
|
raw
history blame
386 Bytes
metadata
datasets:
  - cerebras/SlimPajama-627B
language:
  - en
Tasks Version Filter n-shot Metric Value Stderr
arc_challenge Yaml none 25 acc 0.1775 ± 0.0112
none 25 acc_norm 0.2133 ± 0.0120
truthfulqa_mc2 Yaml none 0 acc 0.4457 ± 0.0152