# General Performance on MMLU, HellaSwag, TruthfulQA, GSM8K, MATH, MBPP, HumanEval

Source: [General Performance on MMLU, HellaSwag, TruthfulQA, GSM8K, MATH, MBPP, HumanEval](<https://www.sota2.com/research/sota/general-performance-on-mmlu-hellaswag-truthfulqa-gsm8k-math-mbpp-humaneval>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:35:49.828Z

Aggregate evaluation assessing general performance across a comprehensive suite including MMLU, HellaSwag, TruthfulQA, GSM8K, MATH, MBPP, and HumanEval.

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 | General Performance |
| Dataset (including any stated split) | Aggregated MMLU, HellaSwag, TruthfulQA, GSM8K, MATH, MBPP, HumanEval |
| Metric | Average Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Average Score](<https://www.sota2.com/research/sota/general-performance-on-mmlu-hellaswag-truthfulqa-gsm8k-math-mbpp-humaneval.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Score | Result date | Report paper |
| --- | --- | --- | --- |
| Sens-Merging (DARE) | 40.35 | Not recorded | 68fa770e1e209131cea70ce0 |
| Sens-Merging (Ties-Merging) | 40.22 | Not recorded | 68fa770e1e209131cea70ce0 |
| Sens-Merging (Task Arithmetic) | 40.2 | Not recorded | 68fa770e1e209131cea70ce0 |
| DARE | 40.13 | Not recorded | 68fa770e1e209131cea70ce0 |
| Ties-Merging | 39.89 | Not recorded | 68fa770e1e209131cea70ce0 |
| Math | 37.42 | Not recorded | 68fa770e1e209131cea70ce0 |
| Task Arithmetic | 34.52 | Not recorded | 68fa770e1e209131cea70ce0 |
| Code | 31.21 | Not recorded | 68fa770e1e209131cea70ce0 |
| Chat | 31.18 | Not recorded | 68fa770e1e209131cea70ce0 |

### Sens-Merging (DARE)

Average Score: 40.35. Result ID: 6994730f3a6c3148491d5353.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":true}. This record does not assert independent reproduction.

### Sens-Merging (Ties-Merging)

Average Score: 40.22. Result ID: 6994730f3a6c3148491d5351.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":true}. This record does not assert independent reproduction.

### Sens-Merging (Task Arithmetic)

Average Score: 40.2. Result ID: 6994730f3a6c3148491d534f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":true}. This record does not assert independent reproduction.

### DARE

Average Score: 40.13. Result ID: 6994730f3a6c3148491d5352.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":false}. This record does not assert independent reproduction.

### Ties-Merging

Average Score: 39.89. Result ID: 6994730f3a6c3148491d5350.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":false}. This record does not assert independent reproduction.

### Math

Average Score: 37.42. Result ID: 6994730f3a6c3148491d534c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B"}. This record does not assert independent reproduction.

### Task Arithmetic

Average Score: 34.52. Result ID: 6994730f3a6c3148491d534e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B","Use Sens":false}. This record does not assert independent reproduction.

### Code

Average Score: 31.21. Result ID: 6994730f3a6c3148491d534d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B"}. This record does not assert independent reproduction.

### Chat

Average Score: 31.18. Result ID: 6994730f3a6c3148491d534b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa770e1e209131cea70ce0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701f1e209131cea492fe.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Backbone":"LLaMA2-13B"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
