Hi, I am trying to fine-tune to make function call predictions better on Meta-Llama-3.1-8B-Instruct. To do that I created a dataset and applied steps regarding to Fine-Tuning Llama-3.1-8B for Function Calling using LoRA | by Gautam Chutani | Medium blog. As a result I can see function name and para…

I think this phenomenon is what is known as “catastrophic forgetting,” but I don’t think there is anything particularly wrong with your method… Perhaps the learning rate is too high, or something like that? [image] Let’s Fine-Tune Your Model for Function-Calling - Hugging Face Agents Cours…

Fine-tune for function call on Meta-Llama-3.1-8B-Instruct

🤗Transformers

system Closed July 15, 2025, 2:11am 7

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.

Topic		Replies	Views
Bad Performance Finetuning Llama Chat and Instruct Models on GSM8K Beginners	5	1116	December 5, 2024
Performance problems with finetuned model (Llama 2 7B based) Beginners	3	687	June 10, 2024
Model Fine Tuning using Llama-2-7b-chat-hf not working for text-to-SQL task Beginners	0	303	June 14, 2024
Fine-tuning don't work / bad results Beginners	5	1692	January 15, 2025
Finetuning Meta-Llama-3.1-8B using PEFT Models	4	3462	February 1, 2025

Fine-tune for function call on Meta-Llama-3.1-8B-Instruct

Related topics