Update README.md
Browse files
README.md
CHANGED
|
@@ -31,6 +31,18 @@ It achieves the following results on the evaluation set:
|
|
| 31 |
- Logits/generated: -2.6844
|
| 32 |
- Logits/real: -2.7197
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
## Model description
|
| 35 |
|
| 36 |
More information needed
|
|
|
|
| 31 |
- Logits/generated: -2.6844
|
| 32 |
- Logits/real: -2.7197
|
| 33 |
|
| 34 |
+
|
| 35 |
+
## MT-Bench results
|
| 36 |
+
|
| 37 |
+
| Model | 1st Turn Score | 2nd Turn Score | Average Score |
|
| 38 |
+
|-------------------------|----------------|----------------|---------------|
|
| 39 |
+
| zephyr-7b-sft-full | 6.6625 | 6.0250 | 6.34375 |
|
| 40 |
+
| zephyr-7b-spin-iter0-v0 | 6.64375 | 6.1750 | 6.409375 |
|
| 41 |
+
| zephyr-7b-spin-iter1-v0 | 6.90625 | 6.3000 | 6.603125 |
|
| 42 |
+
| zephyr-7b-spin-iter2-v0 | **7.1375** | 6.3125 | 6.725000 |
|
| 43 |
+
| zephyr-7b-spin-iter3-v0 | 7.09375 | **6.4500** | **6.771875** |
|
| 44 |
+
|
| 45 |
+
|
| 46 |
## Model description
|
| 47 |
|
| 48 |
More information needed
|