RLTasksby ulam.ai
Models / AI2

AI2 / MODEL EVIDENCE

Tulu-2-70b-dpo

1 benchmarks · 12 observations · 1 sources

Scores at a glance

1 benchmarks · large dot: Tulu-2-70b-dpo; small dots: closest models where reported

CoCoNot (Contextual Non-Compliance) ↗12%
7.641.3

Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Incomplete · compliance rate · reported model comparison ↗

11 more reported results

7.6% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Unsupported · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

1.4% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Indeterminate · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

8.1% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Safety · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

6.1% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Humanizing · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

84.2% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · No system prompt · Contrast set · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

8% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Incomplete · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

12.1% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Unsupported · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

0% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Indeterminate · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

10.6% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Safety · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

1.2% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Humanizing · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

89.5% · Tulu-2-70b-dpo · CoCoNot (Contextual Non-Compliance) · With system prompt · Contrast set · compliance rate · reported model comparison · The Art of Saying No: Contextual Noncompliance in Language Models ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

5.3 percentile-point gap

1 of 1 shared benchmark comparisons

5.3 percentile-point gap

1 of 1 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. The Art of Saying No: Contextual Noncompliance in Language Models — published result table ↗

    Brahman, Faeze, Kumar, Sachin, Balachandran, Vidhisha, Dasigi, Pradeep, Pyatkin, Valentina, Ravichander, Abhilasha et al. · 2024

    All 14 authors

    Brahman, Faeze, Kumar, Sachin, Balachandran, Vidhisha, Dasigi, Pradeep, Pyatkin, Valentina, Ravichander, Abhilasha, Wiegreffe, Sarah, Dziri, Nouha, Chandu, Khyathi, Hessel, Jack, Tsvetkov, Yulia, Smith, Noah A., Choi, Yejin, Hajishirzi, Hannaneh

Names used in source tables

Tulu-2-70b-dpo