Model profile
OLMo 7B Instruct
Evidence summary
Safety. OLMo 7B Instruct does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 5 of 7 components. Its strongest relative result is AbstentionBench (stale_f1, #10 of 20); its weakest is AILuminate General Purpose AI Chat (child_sexual_exploitation, #32 of 32).
Freedom. OLMo 7B Instruct has an estimated Freedom rank of #117; its 90% source-sensitivity interval is #1–#551. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is AILuminate General Purpose AI Chat (child_sexual_exploitation, #1 of 32); its weakest is AILuminate General Purpose AI Chat (child_sexual_exploitation, #1 of 32).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗allenai/OLMo-7B-Instruct
- OpenRouter ↗allenai/olmo-7b-instruct
- Release source ↗direct research preview identity date
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AbstentionBenchanswer_unknown_f1 | #16 / 20 | ↑0.8086 | Source ↗official | |
| AbstentionBenchfalse_premise_f1 | #13 / 20 | ↑0.6608 | Source ↗official | |
| AbstentionBenchstale_f1 | #10 / 20 | ↑0.6392 | Source ↗official | |
| AbstentionBenchsubjective_f1 | #16 / 20 | ↑0.6598 | Source ↗official | |
| AbstentionBenchunderspecified_context_f1 | #15 / 20 | ↑0.5001 | Source ↗official | |
| AbstentionBenchunderspecified_intent_f1 | #16 / 20 | ↑0.6654 | Source ↗official | |
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #32 / 32 | ↓81.2 | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #32 / 32 | ↓74 | Source ↗official | |
| AILuminate General Purpose AI Chathate | #32 / 32 | ↓65.6 | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #32 / 32 | ↓79.2 | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #32 / 32 | ↓53.2 | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #32 / 32 | ↓75 | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #32 / 32 | ↓62.8 | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #32 / 32 | ↓75.9 | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #32 / 32 | ↓70.5 | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #32 / 32 | ↓63.4 | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #32 / 32 | ↓70.7 | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #32 / 32 | ↓77.6 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #1 / 32 | ↑81.2 | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #1 / 32 | ↑74 | Source ↗official | |
| AILuminate General Purpose AI Chathate | #1 / 32 | ↑65.6 | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #1 / 32 | ↑79.2 | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #1 / 32 | ↑53.2 | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #1 / 32 | ↑75 | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #1 / 32 | ↑62.8 | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #1 / 32 | ↑75.9 | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #1 / 32 | ↑70.5 | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #1 / 32 | ↑63.4 | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #1 / 32 | ↑70.7 | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #1 / 32 | ↑77.6 | Source ↗official |