readme: add more training details
Browse files
README.md
CHANGED
@@ -99,7 +99,7 @@ print(generated_text)
|
|
99 |
Notice: this model integration is heavily under development. And in the process of finding good hyper-parameters.
|
100 |
Also downstream experiments are coming very soon.
|
101 |
|
102 |
-
Unfortunately, there are nan's occuring in the training:
|
103 |
|
104 |
![Training Loss](training-loss.png)
|
105 |
|
|
|
99 |
Notice: this model integration is heavily under development. And in the process of finding good hyper-parameters.
|
100 |
Also downstream experiments are coming very soon.
|
101 |
|
102 |
+
Unfortunately, there are nan's occuring in the training (after 7h 33m 14s of training on a single RTX 4090):
|
103 |
|
104 |
![Training Loss](training-loss.png)
|
105 |
|