NotASI commited on
Commit
b24f252
·
verified ·
1 Parent(s): 1f203a6

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +4 -2
app.py CHANGED
@@ -7,7 +7,7 @@ client = Cerebras(
7
  )
8
 
9
  TTILE = """
10
- <h1 align="center">Try the world's fastest inference 🚀</h1>
11
  """
12
  NOTICE = """
13
  Current only support Llama3.1 8B and Llama 3.1 70B.
@@ -17,11 +17,12 @@ def respond(
17
  message,
18
  history: list[tuple[str, str]],
19
  model_id,
 
20
  max_tokens,
21
  temperature,
22
  top_p,
23
  ):
24
- messages = []
25
 
26
  for val in history:
27
  if val[0]:
@@ -56,6 +57,7 @@ chatbot = gr.ChatInterface(
56
  value="llama3.1-8b",
57
  label="Models"
58
  ),
 
59
  gr.Slider(minimum=1, maximum=8192, value=4096, step=1, label="Max new tokens"),
60
  gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
61
  gr.Slider(
 
7
  )
8
 
9
  TTILE = """
10
+ <h1 align="center">🚀 Try the world's fastest inference by Cerebras ⚡</h1>
11
  """
12
  NOTICE = """
13
  Current only support Llama3.1 8B and Llama 3.1 70B.
 
17
  message,
18
  history: list[tuple[str, str]],
19
  model_id,
20
+ system_message,
21
  max_tokens,
22
  temperature,
23
  top_p,
24
  ):
25
+ messages = [{"role": "system", "content": system_message}]
26
 
27
  for val in history:
28
  if val[0]:
 
57
  value="llama3.1-8b",
58
  label="Models"
59
  ),
60
+ gr.Textbox(value="You are a friendly assistant.", label="System message"),
61
  gr.Slider(minimum=1, maximum=8192, value=4096, step=1, label="Max new tokens"),
62
  gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
63
  gr.Slider(