I use “gpt-3.5-turbo-0613”.
Following error occured:
“Error: This model’s maximum context length is 4097 tokens. However, you requested 7409 tokens (1345 in the messages, 64 in the functions, and 6000 in the completion)”.
I required response within “30 words”.
Tell me how solve this problem. This error never happen with “gpt-3.5-turbo”.
Would you mind to post the messages here?
Seems really strange.
What’s the max_tokens of the request?
i set max_tokens 4000 or did not set the value.
max_tokens = 4000 equals to completion value.
I set 6000, completion value became 6000.
1 Like
Sorry, I prompt by Japanese.
I tried to paste the github link. But I couldn’t send the message.
sota1111/tmp
That might cause the error.
Hiragana: Unicode: 3040-309F
Katakana: Unicode: 30A0–30FF
I mean, I saw something alike in chinese translation responses. I guess GPT-3 can’t identify words in japanese, or at least not reliable enough.
And cause of that max_tokens is taken.
You’ve put a 4k limit and the answer is transscripted to unicode?
problem solved. max_tokens equals to completion value.
Thank you @jochenschultz and @kjordan 
1 Like