chenghenry
commited on
Commit
•
3db0199
1
Parent(s):
8bb1eef
Update README.md
Browse files
README.md
CHANGED
@@ -2,4 +2,39 @@
|
|
2 |
license: gemma
|
3 |
library_name: transformers
|
4 |
base_model: google/gemma-1.1-7b-it
|
5 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
license: gemma
|
3 |
library_name: transformers
|
4 |
base_model: google/gemma-1.1-7b-it
|
5 |
+
---
|
6 |
+
|
7 |
+
## Usage (llama-cli with GPU):
|
8 |
+
```
|
9 |
+
llama-cli -m ./gemma-1.1-7b-it-Q6_K.gguf -ngl 100 --temp 0 --repeat-penalty 1.0 --color -p "Why is the sky blue?"
|
10 |
+
```
|
11 |
+
|
12 |
+
## Usage (llama-cli with CPU):
|
13 |
+
```
|
14 |
+
llama-cli -m ./gemma-1.1-7b-it-Q6_K.gguf --temp 0 --repeat-penalty 1.0 --color -p "Why is the sky blue?"
|
15 |
+
```
|
16 |
+
|
17 |
+
## Usage (llama-cpp-python via Hugging Face Hub):
|
18 |
+
```
|
19 |
+
from llama_cpp import Llama
|
20 |
+
|
21 |
+
llm = Llama.from_pretrained(
|
22 |
+
repo_id="chenghenry/gemma-1.1-7b-it-GGUF",
|
23 |
+
filename="gemma-1.1-7b-it-Q6_K.gguf",
|
24 |
+
n_ctx=8192,
|
25 |
+
n_batch=2048,
|
26 |
+
n_gpu_layers=100,
|
27 |
+
verbose=False,
|
28 |
+
chat_format="gemma"
|
29 |
+
)
|
30 |
+
|
31 |
+
prompt = "Why is the sky blue?"
|
32 |
+
|
33 |
+
messages = [{"role": "user", "content": prompt}]
|
34 |
+
response = llm.create_chat_completion(
|
35 |
+
messages=messages,
|
36 |
+
repeat_penalty=1.0,
|
37 |
+
temperature=0)
|
38 |
+
|
39 |
+
print(response["choices"][0]["message"]["content"])
|
40 |
+
```
|