Update app.py
Browse files
app.py
CHANGED
@@ -17,6 +17,16 @@ client = OpenAI(
|
|
17 |
api_key=ACCESS_TOKEN,
|
18 |
)
|
19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
def respond(
|
21 |
message,
|
22 |
history: list[tuple[str, str]],
|
@@ -25,7 +35,7 @@ def respond(
|
|
25 |
temperature,
|
26 |
top_p,
|
27 |
):
|
28 |
-
messages = [{"role": "system", "content":
|
29 |
|
30 |
for val in history:
|
31 |
if val[0]:
|
@@ -37,7 +47,7 @@ def respond(
|
|
37 |
|
38 |
response = ""
|
39 |
|
40 |
-
for message in
|
41 |
model="meta-llama/Meta-Llama-3.1-8B-Instruct",
|
42 |
max_tokens=max_tokens,
|
43 |
stream=True,
|
@@ -53,7 +63,7 @@ def respond(
|
|
53 |
demo = gr.ChatInterface(
|
54 |
respond,
|
55 |
additional_inputs=[
|
56 |
-
gr.Textbox(value=
|
57 |
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
|
58 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
59 |
gr.Slider(
|
@@ -63,9 +73,9 @@ demo = gr.ChatInterface(
|
|
63 |
step=0.05,
|
64 |
label="Top-P",
|
65 |
),
|
66 |
-
|
67 |
],
|
68 |
css=css
|
69 |
)
|
|
|
70 |
if __name__ == "__main__":
|
71 |
demo.launch()
|
|
|
17 |
api_key=ACCESS_TOKEN,
|
18 |
)
|
19 |
|
20 |
+
SYSTEM_PROMPT = """You are an expert assistant who can solve any task using code blobs. You will be given a task to solve as best you can.
|
21 |
+
To do so, you have been given access to a list of tools: these tools are basically Python functions which you can call with code.
|
22 |
+
To solve the task, you must plan forward to proceed in a series of steps, in a cycle of 'Thought:', 'Code:', and 'Observation:' sequences.
|
23 |
+
At each step, in the 'Thought:' sequence, you should first explain your reasoning towards solving the task and the tools that you want to use.
|
24 |
+
Then in the 'Code:' sequence, you should write the code in simple Python. The code sequence must end with '<end_code>' sequence.
|
25 |
+
During each intermediate step, you can use 'print()' to save whatever important information you will then need.
|
26 |
+
These print outputs will then appear in the 'Observation:' field, which will be available as input for the next step.
|
27 |
+
In the end, you have to return a final answer using the `final_answer` tool.
|
28 |
+
"""
|
29 |
+
|
30 |
def respond(
|
31 |
message,
|
32 |
history: list[tuple[str, str]],
|
|
|
35 |
temperature,
|
36 |
top_p,
|
37 |
):
|
38 |
+
messages = [{"role": "system", "content": SYSTEM_PROMPT}]
|
39 |
|
40 |
for val in history:
|
41 |
if val[0]:
|
|
|
47 |
|
48 |
response = ""
|
49 |
|
50 |
+
for message in client.chat.completions.create(
|
51 |
model="meta-llama/Meta-Llama-3.1-8B-Instruct",
|
52 |
max_tokens=max_tokens,
|
53 |
stream=True,
|
|
|
63 |
demo = gr.ChatInterface(
|
64 |
respond,
|
65 |
additional_inputs=[
|
66 |
+
gr.Textbox(value=SYSTEM_PROMPT, label="System message"),
|
67 |
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
|
68 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
69 |
gr.Slider(
|
|
|
73 |
step=0.05,
|
74 |
label="Top-P",
|
75 |
),
|
|
|
76 |
],
|
77 |
css=css
|
78 |
)
|
79 |
+
|
80 |
if __name__ == "__main__":
|
81 |
demo.launch()
|