Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on GSM8K (Overall, Prune)
Loading...
22.079
Overall Accuracy
SLM
3.94036
8.64943
13.3585
18.06757
Feb 24, 2025
Overall Accuracy
Prune Accuracy
Updated 26d ago
Evaluation Results
Method
Method
Links
Overall Accuracy
Prune Accuracy
SLM
Backbone=Llama3-70B-In...
2025.02
22.079
15
H2O
Backbone=Llama3-70B-In...
2025.02
19.062
11.4
Llama3 (Full)
Backbone=Llama3-70B-In...
2025.02
17.504
-
SnapKV
Backbone=Llama3-70B-In...
2025.02
16.517
11.7
ReFreeKV
Backbone=Llama3-70B-In...
2025.02
15.975
15.4
SLM
Backbone=Llama3-8B-Ins...
2025.02
6.071
0.6
SnapKV
Backbone=Llama3-8B-Ins...
2025.02
5.874
2.1
Llama3 (Full)
Backbone=Llama3-8B-Ins...
2025.02
4.693
-
H2O
Backbone=Llama3-8B-Ins...
2025.02
4.686
2
ReFreeKV
Backbone=Llama3-8B-Ins...
2025.02
4.638
3.4
Feedback
Search any
task
Search any
task