← Models

Model profile

Llama 3.1 Tulu 3 8B Ppo Rlvf

2024-11-20release date
1eval lineages

Evidence summary

Published evidence spans 1 evals and 2 of 7 behavior components. Its strongest relative result is AbstentionBench (stale_f1, #2 of 20); its weakest is AbstentionBench (underspecified_context_f1, #19 of 20).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

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-evalRankValueDistributionBetterSource
AbstentionBenchanswer_unknown_f1#5 / 200.8968↑ higherSource ↗official
AbstentionBenchfalse_premise_f1#15 / 200.6531↑ higherSource ↗official
AbstentionBenchstale_f1#2 / 200.7216↑ higherSource ↗official
AbstentionBenchsubjective_f1#8 / 200.7493↑ higherSource ↗official
AbstentionBenchunderspecified_context_f1#19 / 200.4319↑ higherSource ↗official
AbstentionBenchunderspecified_intent_f1#5 / 200.7569↑ higherSource ↗official