llama-anon
commited on
Commit
•
3bd795d
1
Parent(s):
f1d550d
Update README.md
Browse files
README.md
CHANGED
@@ -5,7 +5,7 @@ license: agpl-3.0
|
|
5 |
|
6 |
LLaMA-13B merged with Instruct-13B weights, unlike the bare weights it does not output gibberish.
|
7 |
|
8 |
-
Quantized to 4bit (
|
9 |
|
10 |
Feel free to donate:
|
11 |
|
|
|
5 |
|
6 |
LLaMA-13B merged with Instruct-13B weights, unlike the bare weights it does not output gibberish.
|
7 |
|
8 |
+
Quantized to 4bit (with mmap enabled)
|
9 |
|
10 |
Feel free to donate:
|
11 |
|