# Bias Evaluation on Task 2 Persona F Source: [Bias Evaluation on Task 2 Persona F]() Collection index: [llms.txt]() Data updated: 2026-04-09T23:08:38.769Z Evaluation of model bias using the Task 2 Persona F dataset, reporting the BAD Score. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Bias Evaluation | | Dataset (including any stated split) | Task 2 Persona F | | Metric | BAD Score | | Unit | PERCENT | | Direction | Lower is better | | Results loaded | 25 of 25 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [BAD Score](): Lower is better; unit: PERCENT. ## Recorded results | Method | BAD Score | Result date | Report paper | | --- | --- | --- | --- | | DeepSeek-R1 | 0.3 | Not recorded | 69d732637d0172140ec76969 | | LLaMA-4 | 1 | Not recorded | 69d732637d0172140ec76969 | | DeepSeek-R1 | 1.8 | Not recorded | 69d732637d0172140ec76969 | | Gemma-3n E4B | 2.6 | Not recorded | 69d732637d0172140ec76969 | | GPT-4o | 4.3 | Not recorded | 69d732637d0172140ec76969 | | LLaMA-4 | 4.8 | Not recorded | 69d732637d0172140ec76969 | | Claude 4.o Sonnet | 5.8 | Not recorded | 69d732637d0172140ec76969 | | Gemma-3n E4B | 6.1 | Not recorded | 69d732637d0172140ec76969 | | Gemma-3n E4B | 6.7 | Not recorded | 69d732637d0172140ec76969 | | Claude 4.o Sonnet | 7.1 | Not recorded | 69d732637d0172140ec76969 | | DeepSeek-R1 | 7.5 | Not recorded | 69d732637d0172140ec76969 | | GPT-4o | 7.8 | Not recorded | 69d732637d0172140ec76969 | | Gemma-3n E4B | 8 | Not recorded | 69d732637d0172140ec76969 | | Gemma-3n E4B | 9.1 | Not recorded | 69d732637d0172140ec76969 | | Claude 4.o Sonnet | 9.4 | Not recorded | 69d732637d0172140ec76969 | | LLaMA-4 | 9.8 | Not recorded | 69d732637d0172140ec76969 | | GPT-4o | 10.7 | Not recorded | 69d732637d0172140ec76969 | | Claude 4.o Sonnet | 11.2 | Not recorded | 69d732637d0172140ec76969 | | LLaMA-4 | 11.5 | Not recorded | 69d732637d0172140ec76969 | | DeepSeek-R1 | 12 | Not recorded | 69d732637d0172140ec76969 | | LLaMA-4 | 12.6 | Not recorded | 69d732637d0172140ec76969 | | Claude 4.o Sonnet | 21.1 | Not recorded | 69d732637d0172140ec76969 | | GPT-4o | 22.6 | Not recorded | 69d732637d0172140ec76969 | | GPT-4o | 28.3 | Not recorded | 69d732637d0172140ec76969 | | DeepSeek-R1 | 30.8 | Not recorded | 69d732637d0172140ec76969 | ### DeepSeek-R1 BAD Score: 0.3. Result ID: 69d8316647f34ce924bfd3f0. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P4"}. This record does not assert independent reproduction. ### LLaMA-4 BAD Score: 1. Result ID: 69d8316647f34ce924bfd3f3. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P2"}. This record does not assert independent reproduction. ### DeepSeek-R1 BAD Score: 1.8. Result ID: 69d8316647f34ce924bfd3ee. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P2"}. This record does not assert independent reproduction. ### Gemma-3n E4B BAD Score: 2.6. Result ID: 69d8316647f34ce924bfd3fd. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P2"}. This record does not assert independent reproduction. ### GPT-4o BAD Score: 4.3. Result ID: 69d8316647f34ce924bfd3e9. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P2"}. This record does not assert independent reproduction. ### LLaMA-4 BAD Score: 4.8. Result ID: 69d8316647f34ce924bfd3f4. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P3"}. This record does not assert independent reproduction. ### Claude 4.o Sonnet BAD Score: 5.8. Result ID: 69d8316647f34ce924bfd3fb. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P5"}. This record does not assert independent reproduction. ### Gemma-3n E4B BAD Score: 6.1. Result ID: 69d8316647f34ce924bfd3fe. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P3"}. This record does not assert independent reproduction. ### Gemma-3n E4B BAD Score: 6.7. Result ID: 69d8316647f34ce924bfd3fc. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P1"}. This record does not assert independent reproduction. ### Claude 4.o Sonnet BAD Score: 7.1. Result ID: 69d8316647f34ce924bfd3f7. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P1"}. This record does not assert independent reproduction. ### DeepSeek-R1 BAD Score: 7.5. Result ID: 69d8316647f34ce924bfd3ef. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P3"}. This record does not assert independent reproduction. ### GPT-4o BAD Score: 7.8. Result ID: 69d8316647f34ce924bfd3e8. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P1"}. This record does not assert independent reproduction. ### Gemma-3n E4B BAD Score: 8. Result ID: 69d8316647f34ce924bfd3ff. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P4"}. This record does not assert independent reproduction. ### Gemma-3n E4B BAD Score: 9.1. Result ID: 69d8316647f34ce924bfd400. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P5"}. This record does not assert independent reproduction. ### Claude 4.o Sonnet BAD Score: 9.4. Result ID: 69d8316647f34ce924bfd3f9. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P3"}. This record does not assert independent reproduction. ### LLaMA-4 BAD Score: 9.8. Result ID: 69d8316647f34ce924bfd3f5. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P4"}. This record does not assert independent reproduction. ### GPT-4o BAD Score: 10.7. Result ID: 69d8316647f34ce924bfd3eb. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P4"}. This record does not assert independent reproduction. ### Claude 4.o Sonnet BAD Score: 11.2. Result ID: 69d8316647f34ce924bfd3fa. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P4"}. This record does not assert independent reproduction. ### LLaMA-4 BAD Score: 11.5. Result ID: 69d8316647f34ce924bfd3f2. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P1"}. This record does not assert independent reproduction. ### DeepSeek-R1 BAD Score: 12. Result ID: 69d8316647f34ce924bfd3f1. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P5"}. This record does not assert independent reproduction. ### LLaMA-4 BAD Score: 12.6. Result ID: 69d8316647f34ce924bfd3f6. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P5"}. This record does not assert independent reproduction. ### Claude 4.o Sonnet BAD Score: 21.1. Result ID: 69d8316647f34ce924bfd3f8. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P2"}. This record does not assert independent reproduction. ### GPT-4o BAD Score: 22.6. Result ID: 69d8316647f34ce924bfd3ea. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P3"}. This record does not assert independent reproduction. ### GPT-4o BAD Score: 28.3. Result ID: 69d8316647f34ce924bfd3ec. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P5"}. This record does not assert independent reproduction. ### DeepSeek-R1 BAD Score: 30.8. Result ID: 69d8316647f34ce924bfd3ed. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"P1"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.