Update README.md
Browse files
README.md
CHANGED
|
@@ -81,7 +81,11 @@ The model has substantially improved German skills as demonstrated in RAG evalua
|
|
| 81 |
|
| 82 |
Stay tuned for the next big models employing Spectrum CPT!
|
| 83 |
|
|
|
|
| 84 |
|
|
|
|
|
|
|
|
|
|
| 85 |
|
| 86 |
|
| 87 |
## Evaluation
|
|
|
|
| 81 |
|
| 82 |
Stay tuned for the next big models employing Spectrum CPT!
|
| 83 |
|
| 84 |
+
**NOTE**
|
| 85 |
|
| 86 |
+
For the demo, we are satisfied with the performance of the model.
|
| 87 |
+
For productive use, more German tokens can be trained on the SauerkrautLM-1.5b as required in order to teach the model even firmer German while only having a relative influence on the performance of the model (25% of the layers).
|
| 88 |
+
The SauerkrautLM-1.5b offers an excellent starting point for this.
|
| 89 |
|
| 90 |
|
| 91 |
## Evaluation
|