I have finetuned Helsinki opus NLP model for translation from english to malayalam where the pretrained model size was around 230mb and after i finetuned this model on a small dataset the model’s size rose to 500mb, does anyone knows on why the size becomes twice as large and if there are any optimization techniques to mitigate this problem, thank you in advance.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Dataset parameters to finetune a pretrained translation model on new vocabulary | 0 | 359 | July 5, 2023 | |
Finetuning neox 20b, why is resulting model so small | 1 | 290 | September 19, 2022 | |
When I use Trainer API to train the GLM Model and save this model,I find memory of the finetuned model is twice the size of the original model. What is the reason for this? | 5 | 316 | March 22, 2023 | |
Anyone have idea how we can finetune a model using Trainer API? | 0 | 439 | April 22, 2022 | |
Finetuning a Large Language Model | 0 | 74 | October 23, 2024 |