Commit
•
8dec167
1
Parent(s):
998403c
Update README.md
Browse files
README.md
CHANGED
@@ -35,7 +35,7 @@ You can run this model out to 32K context with alpha_rope set to 1.
|
|
35 |
* I recommend using Quadratic Sampling (i.e. smoothing factor) for creative work. I think this version performs best with a smoothing factor close to 0.2.
|
36 |
* I recommend using Min-P. Experiment to find your best setting. I find this model tolerates high Min-P settings rather nicely, but use whatever floats your boat.
|
37 |
* You can enable dynamic temperature if you want, but that adds yet another variable to consider and I find it's unnecessary with you're already using Min-P and smoothing factor.
|
38 |
-
* If you use Textgen WebUI as your backend, I recommend enabling the DRY
|
39 |
|
40 |
Experiment with any and all of the settings below! What suits my preferences may not suit yours.
|
41 |
|
@@ -466,10 +466,10 @@ dtype: float16
|
|
466 |
---
|
467 |
name: new-dawn-llama3-70b-32K-v1.0
|
468 |
models:
|
469 |
-
- model:
|
470 |
-
- model:
|
471 |
merge_method: slerp
|
472 |
-
base_model:
|
473 |
parameters:
|
474 |
t:
|
475 |
- value: 0.5
|
|
|
35 |
* I recommend using Quadratic Sampling (i.e. smoothing factor) for creative work. I think this version performs best with a smoothing factor close to 0.2.
|
36 |
* I recommend using Min-P. Experiment to find your best setting. I find this model tolerates high Min-P settings rather nicely, but use whatever floats your boat.
|
37 |
* You can enable dynamic temperature if you want, but that adds yet another variable to consider and I find it's unnecessary with you're already using Min-P and smoothing factor.
|
38 |
+
* If you use Textgen WebUI as your backend, I recommend enabling the DRY sampler settings to reduce repititions, otherwise some repitition penalty plus frequency penalty ought to do the trick.
|
39 |
|
40 |
Experiment with any and all of the settings below! What suits my preferences may not suit yours.
|
41 |
|
|
|
466 |
---
|
467 |
name: new-dawn-llama3-70b-32K-v1.0
|
468 |
models:
|
469 |
+
- model: new-dawn-llama3-70b-v0.16-32K
|
470 |
+
- model: new-dawn-llama3-70b-v0.18-32K
|
471 |
merge_method: slerp
|
472 |
+
base_model: new-dawn-llama3-70b-v0.16-32K
|
473 |
parameters:
|
474 |
t:
|
475 |
- value: 0.5
|