continuation token feature | OpenAI | Page 1

rose geode Jan 12, 2023, 5:18 PM

#

I want to take advantage of the continuation token feature. Did this get depreciated? Chat GPT explains that its available using the ctoken?

It gave an example of how to use it below

import openai

# Set the API key
openai.api_key = "YOUR_API_KEY"

# Initial prompt and parameters
prompt = "Write a story about a spaceship that travels to a distant planet."
stop = "The end"

# Make the initial API call and get the continuation token
response = openai.Completion.create(model="text-davinci-002", prompt=prompt, stop=stop)
continuation_token = response["choices"][0]["ctoken"]

# Use the continuation token to continue generating text
response = openai.Completion.create(model="text-davinci-002", prompt=continuation_token, stop=stop)
generated_text = response["choices"][0]["text"]

# Print the generated text
print(generated_text)

But in the response there is no ctoken. I checked a few models and its missing. I cant find any documentation on this either.

rose geode Jan 12, 2023, 5:40 PM

#

The continuation token feature is to overcome the token limit for an API call by keeping the same context and breaking up the reply into multiple API calls.

sly gust Jan 12, 2023, 5:54 PM

#

Chatgpt made it up. There never was a ctoken

verbal questBOT Jan 12, 2023, 5:56 PM

#

@rose geode

DelonMusk's message blocked by AutoMod

We strive to create a positive environment and have implemented additional filters to prevent the use of inappropriate or disruptive language and topics. Thank you for your cooperation.

#continuation token feature