Upload benchmarks.csv with huggingface_hub
#9
by
likhonsheikh - opened
- benchmarks.csv +7 -0
benchmarks.csv
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Benchmark,Base %,Distilled %,Std Dev
|
| 2 |
+
AIME 2024,1.5,35.2,0.8
|
| 3 |
+
MATH-500,25.0,89.1,1.2
|
| 4 |
+
GSM8K,65.0,92.8,0.5
|
| 5 |
+
GPQA Diamond,28.0,45.5,1.5
|
| 6 |
+
LiveCodeBench,15.0,32.5,2.1
|
| 7 |
+
HumanEval,55.0,82.3,1.8
|