Update README.md
#2
by
ETraKoZ
- opened
README.md
CHANGED
@@ -61,7 +61,7 @@ Additional information about the datasets will be included in the Meditron-3 pub
|
|
61 |
| Model Name | MedmcQA | MedQA | PubmedQA | Average |
|
62 |
|------------------------|---------|--------|----------|---------|
|
63 |
| microsoft/phi-4 | 63.11 | 62.77 | 79.00 | 68.29 |
|
64 |
-
| MePhitron
|
65 |
| Difference (MePhitron vs.) | 3.47 | 6.52 | -1.40 | 2.86 |
|
66 |
|
67 |
We evaluated Meditron on medical multiple-choice questions using [lm-harness](https://github.com/EleutherAI/lm-evaluation-harness) for reproducibility.
|
|
|
61 |
| Model Name | MedmcQA | MedQA | PubmedQA | Average |
|
62 |
|------------------------|---------|--------|----------|---------|
|
63 |
| microsoft/phi-4 | 63.11 | 62.77 | 79.00 | 68.29 |
|
64 |
+
| MePhitron (Meditron-3-Phi4-14B) | 66.58 | 69.29 | 77.60 | 71.16 |
|
65 |
| Difference (MePhitron vs.) | 3.47 | 6.52 | -1.40 | 2.86 |
|
66 |
|
67 |
We evaluated Meditron on medical multiple-choice questions using [lm-harness](https://github.com/EleutherAI/lm-evaluation-harness) for reproducibility.
|