Update README.md
Browse files
README.md
CHANGED
|
@@ -61,7 +61,7 @@ Unlike earlier LLMs that demanded rigid prompt engineering, vibe-code interactio
|
|
| 61 |
|
| 62 |
| Tasks |Version| Filter |n-shot| Metric | | gpt-oss-20b-rl | gpt-oss-20 |
|
| 63 |
|-----------------------|------:|----------------|-----:|-----------|---|----:|----:|
|
| 64 |
-
|gpqa_diamond_cot_n_shot| 2|flexible-extract| 5|exact_match|↑ | 0.7633|
|
| 65 |
|humaneval| 1|create_test| 0|pass@1| |0.8452|
|
| 66 |
## Example Usage
|
| 67 |
|
|
|
|
| 61 |
|
| 62 |
| Tasks |Version| Filter |n-shot| Metric | | gpt-oss-20b-rl | gpt-oss-20 |
|
| 63 |
|-----------------------|------:|----------------|-----:|-----------|---|----:|----:|
|
| 64 |
+
|gpqa_diamond_cot_n_shot| 2|flexible-extract| 5|exact_match|↑ | 0.7633| 0.715
|
| 65 |
|humaneval| 1|create_test| 0|pass@1| |0.8452|
|
| 66 |
## Example Usage
|
| 67 |
|