Qwen2.5 7B Instruct 1M

Evidence summary

Safety. Qwen2.5 7B Instruct 1M does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 3 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #140 of 248); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #181 of 248).

Freedom. Qwen2.5 7B Instruct 1M has an estimated Freedom rank of #148; its 90% source-sensitivity interval is #59–#582. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #66 of 248); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #105 of 248).