Update README.md
Browse files
README.md
CHANGED
|
@@ -62,14 +62,14 @@ Debugged vibecoder dataset
|
|
| 62 |
|
| 63 |
### 📊 Model Evaluation Results
|
| 64 |
|
| 65 |
-
| Tasks | Version | Filter | n-shot | Metric | VibeCoder-20b-0.02-
|
| 66 |
-
|
| 67 |
-
| gsm8k_cot | 3
|
| 68 |
-
| humaneval | 1
|
| 69 |
-
| mmlu_college_biology | 1
|
| 70 |
-
| mmlu_HS_computer_science | 1
|
| 71 |
-
| computer_security | 1
|
| 72 |
-
| college_computer_science | 1
|
| 73 |
|
| 74 |
---
|
| 75 |
|
|
|
|
| 62 |
|
| 63 |
### 📊 Model Evaluation Results
|
| 64 |
|
| 65 |
+
| Tasks | Version | Filter | n-shot | Metric | VibeCoder-20b-0.02-D*_RL | gpt-oss-20 | Qwen 3 235B |
|
| 66 |
+
|--------------------------|---------|------------------|--------|----------------|-----------------------------|-------------|--------------|
|
| 67 |
+
| gsm8k_cot | 3 | flexible-extract | 3 | exact_match ↑ | 0.8452 (+0.7667) | 0.78 | 0.82 |
|
| 68 |
+
| humaneval | 1 | create_test | 0 | exact_match ↑ | 0.933 (+0.8) | 0.73 | 0.92 |
|
| 69 |
+
| mmlu_college_biology | 1 | create_test | 0 | exact_match ↑ | 1.000 (+ – ) | — | — |
|
| 70 |
+
| mmlu_HS_computer_science | 1 | create_test | 0 | exact_match ↑ | 1.000 (+0.9) | — | — |
|
| 71 |
+
| computer_security | 1 | none | 2 | acc ↑ | 0.8528 (+0.700) | — | — |
|
| 72 |
+
| college_computer_science | 1 | none | 2 | acc ↑ | 0.8528 (+0.700) | — | — |
|
| 73 |
|
| 74 |
---
|
| 75 |
|