Update README.md
Browse files
README.md
CHANGED
|
@@ -15,7 +15,7 @@ tags:
|
|
| 15 |
|
| 16 |
Original name : Llama_3.x_70b_Dolnemhertulimtess_v1.0
|
| 17 |
|
| 18 |
-
This model is essentially a Llama 3.1 smart brick, to be used in second level merges.
|
| 19 |
|
| 20 |
I finally abandonned the 3 stages "smart merges" (like Smarteaz) because they are dilluting too much the source models used with the merge-stock technique, even if the benches and PPL were good, and it was dilluted furthermore into the level 4/5 merges I was doing afterwards. It was too much of a soup.
|
| 21 |
|
|
|
|
| 15 |
|
| 16 |
Original name : Llama_3.x_70b_Dolnemhertulimtess_v1.0
|
| 17 |
|
| 18 |
+
This model is essentially a Llama 3.1 smart brick based on by a 3.0->3.3 "port", to be used in second level merges.
|
| 19 |
|
| 20 |
I finally abandonned the 3 stages "smart merges" (like Smarteaz) because they are dilluting too much the source models used with the merge-stock technique, even if the benches and PPL were good, and it was dilluted furthermore into the level 4/5 merges I was doing afterwards. It was too much of a soup.
|
| 21 |
|