Problem with training the API | OpenAI | Page 1

hoary fossil Feb 20, 2023, 4:26 PM

#

I created a training data file with 1000 records, used the CLI Data preparation tool to validate the data and then created a fine-tune model. That was yesterday at night and let it loading during the night. Today When I woke up today, saw that the process got interrupted and used the openai api fine_tunes.follow -i to resume it

After it finished, I tested it in the playground, it started giving nonsense answers, talking with itself making no sense at all. Anyone know what I am doing wrong?

WhatsApp_Image_2023-02-20_at_11.13.42.jpeg

torn topaz Feb 20, 2023, 10:19 PM

#

May I see your whole prompt?

#

I think you can add more descriptions to the header of the prompt

fresh lake Feb 22, 2023, 10:26 AM

#

hoary fossil I created a training data file with 1000 records, used the CLI Data preparation ...

using proper stop sequences must fix this

torn topaz Feb 22, 2023, 11:16 AM

#

yes set human: and ai: as stop sequences

cunning laurel Feb 23, 2023, 10:25 AM

#

hello i have an error 400 with the api of chatgpt

#

Capture_decran_2023-02-23_a_11.11.13.png

torn topaz Feb 23, 2023, 10:38 AM

#

cunning laurel hello i have an error 400 with the api of chatgpt

First, there is no chatgpt api, you may want to use text-completion. It seems that you are using imagine genration url not the text completion.
This an example in the document:

curl https://api.openai.com/v1/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{"model": "text-davinci-003", "prompt": "Say this is a test", "temperature": 0, "max_tokens": 7}'

#Problem with training the API