update eval
Browse files- evaluation/intro.txt +3 -2
evaluation/intro.txt
CHANGED
|
@@ -19,8 +19,9 @@ In most papers, 200 candidate program completions are sampled, and pass@1, pass@
|
|
| 19 |
|||||
|
| 20 |
|Codex (25M)| 3.21% | 7.1% | 12.89%|
|
| 21 |
|Codex (300M)| 13.17%| 20.37% | 36.27% |
|
| 22 |
-
|Codex (12B)| 28.81%| 46.81% | 72.31% |
|
| 23 |
-
|
|
|
|
| 24 |
|
| 25 |
We can load HumanEval dataset and pass@k metric from 🤗 [`datasets`](https://huggingface.co/docs/datasets/index)
|
| 26 |
|
|
|
|
| 19 |
|||||
|
| 20 |
|Codex (25M)| 3.21% | 7.1% | 12.89%|
|
| 21 |
|Codex (300M)| 13.17%| 20.37% | 36.27% |
|
| 22 |
+
|Codex (12B)| 28.81%| 46.81% | 72.31% |
|
| 23 |
+
|
| 24 |
+
\\
|
| 25 |
|
| 26 |
We can load HumanEval dataset and pass@k metric from 🤗 [`datasets`](https://huggingface.co/docs/datasets/index)
|
| 27 |
|