Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Mathematical and Code Reasoning on ZeroEval (test)
Loading...
67.85
GSM8K Accuracy
Mamba-Llama3
25.2724
36.3262
47.38
58.4338
Aug 27, 2024
GSM8K Accuracy
CRUX Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
GSM8K Accuracy
CRUX Accuracy
Mamba-Llama3
Attention percentage=50%
2024.08
67.85
27.88
Mamba2-Llama3
Attention percentage=50%
2024.08
59.36
24.88
Falcon Mamba-7B
Type=Instruct-tuned
2024.08
41.32
8.88
Mamba-Llama3
Attention percentage=25%
2024.08
40.64
15.62
RecurrentGemma-9B
Type=Instruct-tuned
2024.08
38.51
26.25
Mamba2-Llama3
Attention percentage=25%
2024.08
38.13
13.25
Mamba2-Llama3
Attention percentage=1...
2024.08
35.03
10.25
Mamba-Llama3
Attention percentage=1...
2024.08
26.91
11.25
Feedback
Search any
task
Search any
task