Update README.md
Browse files
README.md
CHANGED
|
@@ -9,8 +9,6 @@ datasets:
|
|
| 9 |
|
| 10 |
Day 2 RP finetune of Apriel 15B, with several iterative improvements from the first version. In particular, coherence at good temperatures (~.7) should be much higher.
|
| 11 |
|
| 12 |
-
Compared to the first version I merged 20% of the original instruct checkpoint back in to mitigate forgetting and to preserve more of the original model's style.
|
| 13 |
-
|
| 14 |
I also fully converted the model to use the Phi 3 format; this comes at the slight tradeoff of the `<|end|>` tag not always tokenizing exactly the same way in a few niche scenarios.
|
| 15 |
|
| 16 |
Further attempts were made to fix formatting issues with asterisks on the base model.
|
|
|
|
| 9 |
|
| 10 |
Day 2 RP finetune of Apriel 15B, with several iterative improvements from the first version. In particular, coherence at good temperatures (~.7) should be much higher.
|
| 11 |
|
|
|
|
|
|
|
| 12 |
I also fully converted the model to use the Phi 3 format; this comes at the slight tradeoff of the `<|end|>` tag not always tokenizing exactly the same way in a few niche scenarios.
|
| 13 |
|
| 14 |
Further attempts were made to fix formatting issues with asterisks on the base model.
|