Spaces:

allenai
/

ZebraLogic

Running

yuchenlin commited on Jul 11

Commit

57a9cf2

•

1 Parent(s): 1757118

zebra logic bench

Files changed (2) hide show

ZeroEval-main/result_dirs/zebra-grid.summary.json CHANGED Viewed

@@ -285,6 +285,17 @@
     "Total Puzzles": 1000,
     "Reason Lens": "1216.40"
   },
   {
     "Model": "gpt-3.5-turbo-0125",
     "Mode": "greedy",

     "Total Puzzles": 1000,
     "Reason Lens": "1216.40"
   },
+  {
+    "Model": "Meta-Llama-3-8B-Instruct",
+    "Mode": "sampling",
+    "Puzzle Acc": "11.00",
+    "Cell Acc": "26.11",
+    "No answer": "22.30",
+    "Easy Puzzle Acc": "36.79",
+    "Hard Puzzle Acc": "0.97",
+    "Total Puzzles": 1000,
+    "Reason Lens": "1282.40"
+  },
   {
     "Model": "gpt-3.5-turbo-0125",
     "Mode": "greedy",

zebra_banner.png ADDED Viewed