# Text-Davinci-003 Model Not Producing Correct Result

**URL:** <https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600>\
**Category:** API\
**Created:** [February 24, 2023, 2:27pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600 "2023-02-24T14:27:13Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![oucler](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/oucler/32/34213_2.png) [@oucler](https://community.openai.com/u/oucler)\
**Post date:** [February 24, 2023, 2:27pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/1 "2023-02-24T14:27:13Z")

</div>

I am using the model for reproducing step by step solution to a simple math question. For instance, I asked this question: “Show me how to solve the equation, x + 6x = 9.”. The model gave me a wrong answer.

 ![image](https://us1.discourse-cdn.com/openai1/original/3X/e/5/e567b74b1ccc820a5847e8117803988155cf757a.png)

You can go [www.mylearnmate.com](http://www.mylearnmate.com) website to test it out.

I used this function to produce the steps:

def get\_answer(text):  
start\_sequence = “\nAI:”  
restart\_sequence = "\nHuman: "

```
response = openai.Completion.create(
  model="text-davinci-003",
  prompt=f"{text} Show the steps.",
  temperature=0.9,
  max_tokens=150,
  top_p=1,
  frequency_penalty=0,
  presence_penalty=0.6,
  stop=[" Human:", " AI:"]
)
return response.choices[0].text.strip()

```

Please let me know how to improve the model.

Thanks,

Onur

---

<div class="post-metadata">

**Author:** ![PaulBellow](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/paulbellow/32/597962_2.png) [@PaulBellow](https://community.openai.com/u/PaulBellow)\
**Post date:** [February 24, 2023, 3:04pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/2 "2023-02-24T15:04:46Z")

</div>

> [@ChatGPT simple math calculation mistake](https://community.openai.com/t/chatgpt-simple-math-calculation-mistake/62780/8):
>
> Made sense to me. Thanks @ruby_coder

More threads available. The little magnifying glass on top is a search button.

Hope this helps!

---

<div class="post-metadata">

**Author:** ![oucler](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/oucler/32/34213_2.png) [@oucler](https://community.openai.com/u/oucler)\
**Post date:** [February 24, 2023, 4:28pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/3 "2023-02-24T16:28:58Z")

</div>

I am sorry the thread doesn’t apply to my case because I don’t have a ChatGPT API which is not available yet.

---

<div class="post-metadata">

**Author:** ![PaulBellow](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/paulbellow/32/597962_2.png) [@PaulBellow](https://community.openai.com/u/PaulBellow)\
**Post date:** [February 24, 2023, 4:52pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/4 "2023-02-24T16:52:14Z")

</div>

Sorry. I just meant in general this topic has been covered before.

That is, calculations are a known weak point for large language models.

---

<div class="post-metadata">

**Author:** ![oucler](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/oucler/32/34213_2.png) [@oucler](https://community.openai.com/u/oucler)\
**Post date:** [February 24, 2023, 5:26pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/5 "2023-02-24T17:26:49Z")

</div>

Is there any other mode that you recommend me to use which is not large language model?

---

<div class="post-metadata">

**Author:** ![PaulBellow](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/paulbellow/32/597962_2.png) [@PaulBellow](https://community.openai.com/u/PaulBellow)\
**Post date:** [February 24, 2023, 5:28pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/6 "2023-02-24T17:28:29Z")

</div>

Not off hand. What are you trying to accomplish?

Where else have you searched so far on your journey?

---

<div class="post-metadata">

**Author:** ![raymonddavey](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/raymonddavey/32/14125_2.png) [@raymonddavey](https://community.openai.com/u/raymonddavey)\
**Post date:** [February 24, 2023, 5:30pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/7 "2023-02-24T17:30:21Z")

</div>

Wolfram alpha is a better alternative. But it is not ChatGPT or GPT related.

---

<div class="post-metadata">

**Author:** ![EricGT](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ericgt/32/20571_2.png) [@EricGT](https://community.openai.com/u/EricGT)\
**Post date:** [February 24, 2023, 5:31pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/8 "2023-02-24T17:31:53Z")

</div>

I was getting the details while @raymonddavey was posting.

Since the link gets corruped when posting here are the steps.

1. Naviage to [Wolfram|Alpha](https://www.wolframalpha.com/)
2. In text enter ` x + 6x = 9 simplify`
3. Press enter

---

<div class="post-metadata">

**Author:** ![oucler](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/oucler/32/34213_2.png) [@oucler](https://community.openai.com/u/oucler)\
**Post date:** [February 24, 2023, 6:20pm UTC](https://community.openai.com/t/text-davinci-003-model-not-producing-correct-result/73600/9 "2023-02-24T18:20:26Z")

</div>

WolframAlpha is good but I want to extract information about the question such as question type linear equation or something else. That information is used later in video generations. WolframAlpha is not flexible but I might need to use two APIs: GPT and WolframAlpha that might resolve my issue.

Thanks for the replies.