Bias Evaluation on CrowS-Pairs (pct-stereotype)
51.25Pct StereotypeQwen 3 0.6B - LFT w. SH (baseline 2)
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen 3 0.6B - LFT w. SH (baseline 2)Model=Qwen 3 0.6B, Variant=LFT w. SH (baseline 2)2025.12 | 51.25 | |
| Qwen 3 0.6B - Pretrained model (baseline 1)Model=Qwen 3 0.6B, Variant=Pretrained model (baseline 1)2025.12 | 52.5 | |
| Qwen 3 0.6B - LFT w. SH-Dgender (GC-CDA)Model=Qwen 3 0.6B, Variant=LFT w. SH-Dgender (GC-CDA)2025.12 | 53.75 | |
| Qwen 3 0.6B - LFT w. SH-N (baseline 3)Model=Qwen 3 0.6B, Variant=LFT w. SH-N (baseline 3)2025.12 | 55 | |
| Qwen 3 0.6B - LFT w. SH-Dgender(BaseCDA)Model=Qwen 3 0.6B, Variant=LFT w. SH-Dgender(BaseCDA)2025.12 | 56.56 | |
| Llama 3.1 8B - LFT w. SH (baseline 2)Model=Llama 3.1 8B, Variant=LFT w. SH (baseline 2)2025.12 | 60.31 | |
| Llama 3.2 1B - LFT w. SH-Dgender (BaseCDA)Model=Llama 3.2 1B, Variant=LFT w. SH-Dgender (BaseCDA)2025.12 | 61.25 | |
| Llama 3.2 1B - LFT w. SH-Dgender (GC-CDA)Model=Llama 3.2 1B, Variant=LFT w. SH-Dgender (GC-CDA)2025.12 | 61.56 | |
| Llama 3.2 1B - LFT w. SH-N (baseline 3)Model=Llama 3.2 1B, Variant=LFT w. SH-N (baseline 3)2025.12 | 62.5 | |
| Llama 3.1 8B - LFT w. SH-Dgender (GC-CDA)Model=Llama 3.1 8B, Variant=LFT w. SH-Dgender (GC-CDA)2025.12 | 62.5 | |
| Llama 3.1 8B - LFT w. SH-Dgender (BaseCDA)Model=Llama 3.1 8B, Variant=LFT w. SH-Dgender (BaseCDA)2025.12 | 63.12 | |
| Llama 3.2 1B - Pretrained model (baseline 1)Model=Llama 3.2 1B, Variant=Pretrained model (baseline 1)2025.12 | 63.75 | |
| Llama 3.1 8B - Pretrained model (baseline 1)Model=Llama 3.1 8B, Variant=Pretrained model (baseline 1)2025.12 | 64.06 | |
| Llama 3.1 8B - LFT w. SH-N (baseline 3)Model=Llama 3.1 8B, Variant=LFT w. SH-N (baseline 3)2025.12 | 64.06 | |
| Llama 3.2 1B - LFT w. SH (baseline 2)Model=Llama 3.2 1B, Variant=LFT w. SH (baseline 2)2025.12 | 64.37 |