|
|
# Geschlossene Domänen-Fragenbeantwortung mit LLMs
|
|
|
|
|
|
import { Tabs, Tab } from 'nextra/components';
|
|
|
import { Callout } from 'nextra/components';
|
|
|
|
|
|
## Hintergrund
|
|
|
|
|
|
Der folgende Prompt testet die Fähigkeiten eines LLM, geschlossene Domain-Fragen zu beantworten, was das Beantworten von Fragen zu einem spezifischen Thema oder Bereich beinhaltet.
|
|
|
|
|
|
<Callout type="warning" emoji="⚠️">
|
|
|
Beachten Sie, dass aufgrund der herausfordernden Natur der Aufgabe
|
|
|
LLMs wahrscheinlich Halluzinationen erzeugen, wenn sie keine
|
|
|
Kenntnisse über die Frage haben.
|
|
|
</Callout>
|
|
|
|
|
|
## Prompt
|
|
|
|
|
|
```markdown
|
|
|
Patientenfakten:
|
|
|
|
|
|
- 20 Jahre alte Frau
|
|
|
- mit einer Vorgeschichte von Anorexia nervosa und Depression
|
|
|
- Blutdruck 100/50, Puls 50, Größe 5’5’’
|
|
|
- von ihrem Ernährungsberater überwiesen, aber leugnet ihre Krankheit
|
|
|
- berichtet, gut zu essen, ist aber stark untergewichtig
|
|
|
|
|
|
Bitte schreibe die oben genannten Daten in eine medizinische Notiz um, unter ausschließlicher Verwendung der oben genannten Informationen.
|
|
|
```
|
|
|
|
|
|
## Code / API
|
|
|
|
|
|
<Tabs items={['GPT-4 (OpenAI)', 'Mixtral MoE 8x7B Instruct (Fireworks)']}>
|
|
|
<Tab>
|
|
|
|
|
|
```python
|
|
|
from openai import OpenAI
|
|
|
client = OpenAI()
|
|
|
|
|
|
response = client.chat.completions.create(
|
|
|
model="gpt-4",
|
|
|
messages=[
|
|
|
{
|
|
|
"role": "user",
|
|
|
"content": "Patient’s facts:\n- 20 year old female\n- with a history of anerxia nervosa and depression\n- blood pressure 100/50, pulse 50, height 5’5’’\n- referred by her nutrionist but is in denial of her illness\n- reports eating fine but is severely underweight\n\nPlease rewrite the data above into a medical note, using exclusively the information above."
|
|
|
}
|
|
|
],
|
|
|
temperature=1,
|
|
|
max_tokens=500,
|
|
|
top_p=1,
|
|
|
frequency_penalty=0,
|
|
|
presence_penalty=0
|
|
|
)
|
|
|
```
|
|
|
</Tab>
|
|
|
|
|
|
<Tab>
|
|
|
```python
|
|
|
import fireworks.client
|
|
|
fireworks.client.api_key = "<FIREWORKS_API_KEY>"
|
|
|
completion = fireworks.client.ChatCompletion.create(
|
|
|
model="accounts/fireworks/models/mixtral-8x7b-instruct",
|
|
|
messages=[
|
|
|
{
|
|
|
"role": "user",
|
|
|
"content": "Patient’s facts:\n- 20 year old female\n- with a history of anerxia nervosa and depression\n- blood pressure 100/50, pulse 50, height 5’5’’\n- referred by her nutrionist but is in denial of her illness\n- reports eating fine but is severely underweight\n\nPlease rewrite the data above into a medical note, using exclusively the information above.",
|
|
|
}
|
|
|
],
|
|
|
stop=["<|im_start|>","<|im_end|>","<|endoftext|>"],
|
|
|
stream=True,
|
|
|
n=1,
|
|
|
top_p=1,
|
|
|
top_k=40,
|
|
|
presence_penalty=0,
|
|
|
frequency_penalty=0,
|
|
|
prompt_truncate_len=1024,
|
|
|
context_length_exceeded_behavior="truncate",
|
|
|
temperature=0.9,
|
|
|
max_tokens=4000
|
|
|
)
|
|
|
```
|
|
|
</Tab>
|
|
|
|
|
|
</Tabs>
|
|
|
|
|
|
## Referenz
|
|
|
|
|
|
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://arxiv.org/abs/2303.12712) (13. April 2023)
|