# Common Sense Reasoning on WinoGrande 2021 (Acc\_norm) Source: [Common Sense Reasoning on WinoGrande 2021 (Acc\_norm)]() Collection index: [llms.txt]() Data updated: 2026-06-08T07:32:23.985Z Evaluation of model performance on the WinoGrande 2021 dataset for common sense reasoning, measured using normalized accuracy (Acc\_norm). 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 | Common Sense Reasoning | | Dataset (including any stated split) | WinoGrande 2021 | | Metric | Normalized Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 26 of 26 | | 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 - [Normalized Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Normalized Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Baseline | 78.06 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 73.72 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 73.32 | Not recorded | 6a264c6490966ade919be87b | | Baseline | 70.8 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 70.64 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 69.85 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 67.17 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 65.98 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 65.9 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 65.82 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 64.09 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 63.3 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 62.43 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 62.12 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 60.85 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 60.06 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 59.98 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 59.27 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 59.19 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 55.17 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 54.85 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 54.38 | Not recorded | 6a264c6490966ade919be87b | | ASVD+ | 51.85 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale (KD) | 51.78 | Not recorded | 6a264c6490966ade919be87b | | SVD-LLM | 51.14 | Not recorded | 6a264c6490966ade919be87b | | SigmaScale | 50.91 | Not recorded | 6a264c6490966ade919be87b | ### Baseline Normalized Accuracy: 78.06. Result ID: 6a266f37fbc373592c22d533. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"1.00x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 73.72. Result ID: 6a266f37fbc373592c22d537. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 73.32. Result ID: 6a266f37fbc373592c22d536. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### Baseline Normalized Accuracy: 70.8. Result ID: 6a266f37fbc373592c22d540. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"1.00x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 70.64. Result ID: 6a266f37fbc373592c22d535. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 69.85. Result ID: 6a266f37fbc373592c22d541. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 67.17. Result ID: 6a266f37fbc373592c22d545. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 65.98. Result ID: 6a266f37fbc373592c22d534. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 65.9. Result ID: 6a266f37fbc373592c22d543. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 65.82. Result ID: 6a266f37fbc373592c22d544. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 64.09. Result ID: 6a266f37fbc373592c22d53b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 63.3. Result ID: 6a266f37fbc373592c22d542. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.90x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 62.43. Result ID: 6a266f37fbc373592c22d53a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 62.12. Result ID: 6a266f37fbc373592c22d549. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 60.85. Result ID: 6a266f37fbc373592c22d548. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 60.06. Result ID: 6a266f37fbc373592c22d538. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 59.98. Result ID: 6a266f37fbc373592c22d547. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 59.27. Result ID: 6a266f37fbc373592c22d539. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 59.19. Result ID: 6a266f37fbc373592c22d546. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.75x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 55.17. Result ID: 6a266f37fbc373592c22d54a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 54.85. Result ID: 6a266f37fbc373592c22d54b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 54.38. Result ID: 6a266f37fbc373592c22d54c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### ASVD+ Normalized Accuracy: 51.85. Result ID: 6a266f37fbc373592c22d53d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale (KD) Normalized Accuracy: 51.78. Result ID: 6a266f37fbc373592c22d53f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SVD-LLM Normalized Accuracy: 51.14. Result ID: 6a266f37fbc373592c22d53c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. ### SigmaScale Normalized Accuracy: 50.91. Result ID: 6a266f37fbc373592c22d53e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B Instruct","Compression Ratio":"0.50x","Zero-shot":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.