Update README.md
#10
by
Abdalgader
- opened
README.md
CHANGED
|
@@ -67,7 +67,10 @@ print(response[0]['generated_text'])
|
|
| 67 |
<br>
|
| 68 |
|
| 69 |
## Benchmarks
|
| 70 |
-
We report in the following table our internal pipeline benchmarks
|
|
|
|
|
|
|
|
|
|
| 71 |
|
| 72 |
|
| 73 |
|
|
|
|
| 67 |
<br>
|
| 68 |
|
| 69 |
## Benchmarks
|
| 70 |
+
We report in the following table our internal pipeline benchmarks.
|
| 71 |
+
- We use [lm-evaluation harness](https://github.com/EleutherAI/lm-evaluation-harness).
|
| 72 |
+
- We report **raw scores**.
|
| 73 |
+
- We use same batch-size across all models.
|
| 74 |
|
| 75 |
|
| 76 |
|