Update README.md
Browse files
README.md
CHANGED
|
@@ -16,9 +16,16 @@ Dev eval at CS-HellaSwag (automatically translated HellaSwag benchmark)
|
|
| 16 |
| csmpt-26.5k steps | 0.4524 |
|
| 17 |
|
| 18 |
|
| 19 |
-
However, we ran validation over the course of training on CS-Hellaswag, and after 100k, the improvements were very noisy if any.
|
| 20 |
The improvement over mistral7b is not significant.
|
| 21 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
# Usage
|
| 24 |
## How to Setup Environment
|
|
|
|
| 16 |
| csmpt-26.5k steps | 0.4524 |
|
| 17 |
|
| 18 |
|
| 19 |
+
However, we ran validation over the course of training on CS-Hellaswag, and after 100k steps, the improvements were very noisy if any.
|
| 20 |
The improvement over mistral7b is not significant.
|
| 21 |
|
| 22 |
+
## Loss
|
| 23 |
+
tbd.
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
## Training Method
|
| 27 |
+
tbd.
|
| 28 |
+
|
| 29 |
|
| 30 |
# Usage
|
| 31 |
## How to Setup Environment
|