Lambent commited on
Commit
31a8311
·
verified ·
1 Parent(s): b240f05

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -22,6 +22,10 @@ This is a merge of pre-trained language models created using [mergekit](https://
22
 
23
  Initial private run of openllm benchmark:
24
 
 
 
 
 
25
  | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
26
  |-------------|------:|------|-----:|--------|---|-----:|---|-----:|
27
  |arc_challenge| 1|none | 25|acc |↑ |0.6527|± |0.0139|
 
22
 
23
  Initial private run of openllm benchmark:
24
 
25
+ Cut it off before the GSM8K, but it definitely looks like the eqbench is an outlier on general capabilities.
26
+
27
+ (Not that it isn't still decently smart.)
28
+
29
  | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
30
  |-------------|------:|------|-----:|--------|---|-----:|---|-----:|
31
  |arc_challenge| 1|none | 25|acc |↑ |0.6527|± |0.0139|