Model profile
Tulu 2 Dpo 70B
2023-11-12release date
1eval lineages
Evidence summary
Published evidence spans 1 evals and 4 of 7 behavior components. Its strongest relative result is SALAD-Bench (mcq_representation_toxicity, #3 of 33); its weakest is SALAD-Bench (attack_enhanced_representation_toxicity, #24 of 33).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗allenai/tulu-2-dpo-70b
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| SALAD-Benchattack_enhanced_human_autonomy_integrity | #24 / 33 | 8.19 | ↑ higher | Source ↗official | |
| SALAD-Benchattack_enhanced_information_safety_harms | #19 / 33 | 8.14 | ↑ higher | Source ↗official | |
| SALAD-Benchattack_enhanced_malicious_use | #24 / 33 | 8.16 | ↑ higher | Source ↗official | |
| SALAD-Benchattack_enhanced_misinformation_harms | #24 / 33 | 8.22 | ↑ higher | Source ↗official | |
| SALAD-Benchattack_enhanced_representation_toxicity | #24 / 33 | 7.35 | ↑ higher | Source ↗official | |
| SALAD-Benchattack_enhanced_socioeconomic_harms | #20 / 33 | 10.82 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_human_autonomy_integrity | #18 / 33 | 94.35 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_information_safety_harms | #17 / 33 | 94.65 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_malicious_use | #19 / 33 | 92.75 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_misinformation_harms | #20 / 33 | 92.96 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_representation_toxicity | #17 / 33 | 88.87 | ↑ higher | Source ↗official | |
| SALAD-Benchbase_socioeconomic_harms | #21 / 33 | 86.49 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_human_autonomy_integrity | #3 / 33 | 73.61 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_information_safety_harms | #4 / 33 | 64.72 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_malicious_use | #3 / 33 | 75.45 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_misinformation_harms | #3 / 33 | 68.1 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_representation_toxicity | #3 / 33 | 68.96 | ↑ higher | Source ↗official | |
| SALAD-Benchmcq_socioeconomic_harms | #3 / 33 | 66.67 | ↑ higher | Source ↗official |