Update README.md
Browse files
README.md
CHANGED
|
@@ -50,14 +50,21 @@ I didn't run any sort of comprehensive set of benchmarks, but here are a couple
|
|
| 50 |
|
| 51 |
### MT-Bench
|
| 52 |
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
|
| 56 |
-
|
|
| 57 |
-
| bagel-dpo-8x7b-v0.2 |
|
| 58 |
-
| bagel-8x7b-v0.2 |
|
| 59 |
-
|
| 60 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
| model | score |
|
| 63 |
| --- | --- |
|
|
|
|
| 50 |
|
| 51 |
### MT-Bench
|
| 52 |
|
| 53 |
+
#### First turn
|
| 54 |
+
|
| 55 |
+
| model | score |
|
| 56 |
+
| --- | --- |
|
| 57 |
+
| bagel-dpo-8x7b-v0.2 | __8.43750__ |
|
| 58 |
+
| bagel-8x7b-v0.2 | 8.05625 |
|
| 59 |
+
|
| 60 |
+
#### Second turn
|
| 61 |
+
|
| 62 |
+
| model | score |
|
| 63 |
+
| --- | --- |
|
| 64 |
+
| bagel-dpo-8x7b-v0.2 | __7.6000__ |
|
| 65 |
+
| bagel-8x7b-v0.2 | 7.1375 |
|
| 66 |
+
|
| 67 |
+
#### Average
|
| 68 |
|
| 69 |
| model | score |
|
| 70 |
| --- | --- |
|