Fizzarolli
commited on
Commit
•
ae651a9
1
Parent(s):
ebaee0a
link quants and ST templates
Browse files
README.md
CHANGED
@@ -17,7 +17,7 @@ experimental "storytelling" and roleplaying finetune of llama 3.
|
|
17 |
|
18 |
|
19 |
## quants
|
20 |
-
|
21 |
|
22 |
## prompting
|
23 |
|
@@ -51,6 +51,7 @@ Hi!</s>
|
|
51 |
Hello!</s>
|
52 |
```
|
53 |
(no, i didn't make these special tokens. yes, that probably makes the model worse. shush.)
|
|
|
54 |
|
55 |
|
56 |
## datasets
|
|
|
17 |
|
18 |
|
19 |
## quants
|
20 |
+
GGUF: https://huggingface.co/Lewdiculous/phencyclidine-8b-v1-GGUF-IQ-Imatrix (thanks @Lewdiculous!)
|
21 |
|
22 |
## prompting
|
23 |
|
|
|
51 |
Hello!</s>
|
52 |
```
|
53 |
(no, i didn't make these special tokens. yes, that probably makes the model worse. shush.)
|
54 |
+
sillytavern templates available for this one! make sure to set both the [context template](https://huggingface.co/Fizzarolli/phencyclidine-8b-v1/blob/main/phencyclidine.json) and [instruct template](https://huggingface.co/Fizzarolli/phencyclidine-8b-v1/blob/main/phencyclidine-instruct.json). any samplers should work, although i recommend Big O from my testing thus far
|
55 |
|
56 |
|
57 |
## datasets
|