Does anyone have experience fine-tuning gpt 4o mini with other languages? I am trying to fine tune it using Korean dataset but it behaves very weirdly. For example, it returns mixed language characters I have never seen or it keeps repeating the same sentence until maximum token limit is reached. Does anyone else have experience in this? What could be the cause?
Related Topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Teaching Gpt4 a new language (Yiddish) | 0 | 381 | January 31, 2024 | |
| Fine-tuning GPT to learn a new coding language | 3 | 3089 | December 24, 2023 | |
| What the theory of GPT Finetune? The result looks not so good | 4 | 767 | October 23, 2023 | |
| Can chatgpt be fluent in a specific language? | 1 | 573 | February 27, 2024 | |
| Is it possible to fine tune GPT4o Mini to produce better quality tabular data? | 0 | 78 | July 26, 2024 |