Model profile
GPT J 6B
2021-06-09release date
2eval lineages
Evidence summary
Published evidence spans 2 evals and 1 of 7 behavior components. Its strongest relative result is HELM Classic RealToxicityPrompts (toxic_prompt_toxic_fraction, #19 of 42); its weakest is ChineseSafe (score, #11 of 22).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗EleutherAI/gpt-j-6B
- 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 |
|---|---|---|---|---|---|
| ChineseSafescore | #11 / 22 | 55.98 | ↑ higher | Source ↗official | |
| HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction | #19 / 42 | 0.01905 | ↓ lower | Source ↗official | |
| HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction | #19 / 42 | 0.0852 | ↓ lower | Source ↗official |