search

JGillette71 / SLLIM

Big Logs Tiny Brains: Q&A on Security Logs with Small Language Models
MIT License
1 stars 0 forks source link

Identify lightweight pre-trained models for fine-tuning. #6

Open JGillette71 opened 1 month ago

JGillette71 commented 1 month ago

As developers, we need to identify at least (2) lightweight pretrained models to fine-tune with our use-case specific dataset. This will allow use to leverage pretraining for model generalization and will expedite the training process to ensure timely completion of the project.

NassosG commented 1 month ago

I have downloaded the IOT dataset and will bring the USB drives to class. it's 16gb :) I have also gathered the Windows event logs dataset, which is 35Gigs and I will also bring to class.

NassosG commented 1 month ago

wrong post here