Update README.md
Browse files
README.md
CHANGED
|
@@ -55,23 +55,23 @@ Performance of Step 3.5 Flash measured across **Reasoning**, **Coding**, and **A
|
|
| 55 |
| # Activated Params | - | 11B | 11B | 15B | 32B | 37B | 37B | 32B |
|
| 56 |
| # Total Params | - | 196B | 196B | 309B | 355B | 671B | 671B | 1043B |
|
| 57 |
| General | | | | | | | | |
|
| 58 |
-
| BBH | 3-shot | 87.
|
| 59 |
-
| MMLU | 5-shot | 83.
|
| 60 |
| MMLU‑Redux | 5-shot | 87.2 | 89.2 | 90.6 | - | 90.0† | 90.4† | 90.2 |
|
| 61 |
-
| MMLU‑Pro | 5-shot |
|
| 62 |
-
| HellaSwag | 10-shot | 91.
|
| 63 |
-
| WinoGrande | 5-shot |
|
| 64 |
-
| GPQA | 5-shot |
|
| 65 |
| SuperGPQA | 5-shot | 41.2 | 41.0 | 41.1 | - | 42.3† | 43.6† | 44.7 |
|
| 66 |
-
| SimpleQA | 5-shot | 28.
|
| 67 |
| Mathematics | | | | | | | | |
|
| 68 |
-
| GSM8K | 8-shot | 88.
|
| 69 |
-
| MATH | 4-shot | 65.
|
| 70 |
| Code | | | | | | | | |
|
| 71 |
-
| HumanEval | 3-shot |
|
| 72 |
-
| MBPP | 3-shot | 79.
|
| 73 |
-
| HumanEval+ | 0-shot |
|
| 74 |
-
| MBPP+ | 0-shot |
|
| 75 |
| MultiPL‑E HumanEval | 0-shot | 63.0 | 67.7 | 59.5 | - | 45.9† | 45.7† | 60.5 |
|
| 76 |
| MultiPL‑E MBPP | 0-shot | 47.9 | 58.0 | 56.7 | - | 52.5† | 50.6† | 58.8 |
|
| 77 |
| Chinese | | | | | | | | |
|
|
|
|
| 55 |
| # Activated Params | - | 11B | 11B | 15B | 32B | 37B | 37B | 32B |
|
| 56 |
| # Total Params | - | 196B | 196B | 309B | 355B | 671B | 671B | 1043B |
|
| 57 |
| General | | | | | | | | |
|
| 58 |
+
| BBH | 3-shot | 87.3 | 88.2 | 88.5 | 86.2 | 88.2† | 88.7† | 88.7 |
|
| 59 |
+
| MMLU | 5-shot | 83.4 | 85.8 | 86.7 | 86.1 | 87.4† | 87.8† | 87.8 |
|
| 60 |
| MMLU‑Redux | 5-shot | 87.2 | 89.2 | 90.6 | - | 90.0† | 90.4† | 90.2 |
|
| 61 |
+
| MMLU‑Pro | 5-shot | 63.6 | 62.3 | 73.2 | - | 58.8† | 62.1† | 69.2 |
|
| 62 |
+
| HellaSwag | 10-shot | 91.0 | 90.2 | 88.5 | 87.1 | 89.2† | 89.4† | 94.6 |
|
| 63 |
+
| WinoGrande | 5-shot | 75.8 | 79.1 | 83.8 | - | 85.9† | 85.6† | 85.3 |
|
| 64 |
+
| GPQA | 5-shot | 43.8 | 41.7 | 43.5* | 33.5* | 43.1* | 37.3* | 43.1* |
|
| 65 |
| SuperGPQA | 5-shot | 41.2 | 41.0 | 41.1 | - | 42.3† | 43.6† | 44.7 |
|
| 66 |
+
| SimpleQA | 5-shot | 28.4 | 31.6 | 20.6 | 30.0 | 26.3† | 27.0† | 35.3 |
|
| 67 |
| Mathematics | | | | | | | | |
|
| 68 |
+
| GSM8K | 8-shot | 88.9 | 88.2 | 92.3 | 87.6 | 91.4† | 91.1† | 92.1 |
|
| 69 |
+
| MATH | 4-shot | 65.7 | 66.8 | 71.0 | 62.6 | 62.6† | 62.5† | 70.2 |
|
| 70 |
| Code | | | | | | | | |
|
| 71 |
+
| HumanEval | 3-shot | 67.0 | 81.1 | 77.4* | 79.8* | 72.5* | 67.7* | 84.8* |
|
| 72 |
+
| MBPP | 3-shot | 79.0 | 79.4 | 81.0* | 81.6* | 74.6* | 75.6* | 89.0* |
|
| 73 |
+
| HumanEval+ | 0-shot | 75.0 | 72.0 | 70.7 | - | 64.6† | 67.7† | - |
|
| 74 |
+
| MBPP+ | 0-shot | 62.4 | 70.6 | 71.4 | - | 72.2† | 69.8† | - |
|
| 75 |
| MultiPL‑E HumanEval | 0-shot | 63.0 | 67.7 | 59.5 | - | 45.9† | 45.7† | 60.5 |
|
| 76 |
| MultiPL‑E MBPP | 0-shot | 47.9 | 58.0 | 56.7 | - | 52.5† | 50.6† | 58.8 |
|
| 77 |
| Chinese | | | | | | | | |
|