Fine-tuning LLM for regression yields low loss during training but not in inference?

surya-narayanan · March 4, 2024, 9:14pm

Hi,

Thanks @nielsr for the guidance - I am also training a regression model, but find when I print my labels inside the trainer, the model is printing the input_ids instead of the regression labels. Any idea where it might be interpreting the task as a causal lm task? I made the changes in task type in the lora config, and my model is an AutoModelForSequenceClassification

Topic		Replies	Views
Reduced inference f1 score with QLoRA finetuned model Intermediate	1	881	September 6, 2023
LLM training loss fluctuation 🤗Transformers	0	951	November 30, 2023
Making fine-tuned LLM model more stable Beginners	3	988	December 30, 2023
FREQUENT LOSS SPIKING in CONTINUE TRAINING LLM 🤗Transformers	2	1059	April 4, 2024
Different results from checkpoint evaluation when loading fine-tuned LLM model Intermediate	5	3239	September 22, 2023

Fine-tuning LLM for regression yields low loss during training but not in inference?

Related topics