Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Language Generation Bias Evaluation on BOLD
Loading...
0.016
Toxicity Score (All)
Mistral-7B
0.01576
0.01738
0.019
0.02062
Jul 18, 2024
Toxicity Score (All)
Toxicity Score (Gender)
Toxicity Score (Race)
Toxicity Score (Religion)
Updated 4d ago
Evaluation Results
Method
Method
Links
Toxicity Score (All)
Toxicity Score (Gender)
Toxicity Score (Race)
Toxicity Score (Religion)
Mistral-7B
Parameters=7B
2024.07
0.016
0.032
0.0146
0.047
Phi-2 + BiasDPO
Parameters=2.7B, Align...
2024.07
0.018
0.03
0.0164
0.0613
StableLM-3B
Parameters=3B
2024.07
0.02
0.033
0.024
0.07
Phi-2
Parameters=2.7B
2024.07
0.02
0.031
0.0144
0.0469
Gemma-2B
Parameters=2B
2024.07
0.022
0.038
0.0139
0.0367
Feedback
Search any
task
Search any
task