Update README.md
Browse files
README.md
CHANGED
|
@@ -13,8 +13,7 @@ Experiment, can DUS be taken one or more steps further?
|
|
| 13 |
- model v03 finetuned on 50k entries from SlimOrca dataset
|
| 14 |
- 12 layers removed from both models, 4 more than in original paper but its 1/4 of all layers(48) as per original paper.
|
| 15 |
- base version of upstage/SOLAR-10.7B-v1.0 used for merge
|
| 16 |
-
- no finetuning done yet, this is just a merge, first step in DUS paper
|
| 17 |
-
- next step, if evaluation proves that its at least as 'smart' as base model, should be finetuning to 'recover' after merge
|
| 18 |
|
| 19 |
-
|
|
|
|
| 20 |
- model performance after finetuning, did it recover initial permofmance loss after merge?
|
|
|
|
| 13 |
- model v03 finetuned on 50k entries from SlimOrca dataset
|
| 14 |
- 12 layers removed from both models, 4 more than in original paper but its 1/4 of all layers(48) as per original paper.
|
| 15 |
- base version of upstage/SOLAR-10.7B-v1.0 used for merge
|
|
|
|
|
|
|
| 16 |
|
| 17 |
+
|
| 18 |
+
### To evaluate
|
| 19 |
- model performance after finetuning, did it recover initial permofmance loss after merge?
|