spankevich/llm-course-hw3-lora

Text Generation • Updated Mar 29 • 2

Note it was used to fine-tune OuteAI/Lite-Oute-1-300M-Instruct for tweet tone classification problem. Default model achieved 0.08 f1-score, while fine-tuned version achieved 0.53 f1-score in less than 8 minutes of fine-tuning on a single A100

spankevich/llm-course-hw3-dora

Text Generation • Updated Mar 29 • 4

Note It was used to fine-tune the same model as LoRA. Fine-tuned version achieved 0.51 f1-score in less than 15 minutes of fine-tuning on a single A100

spankevich/llm-course-hw3-tinyllamma-qlora

Updated Mar 29

Note It was used to fine-tune a bigger model TinyLlama/TinyLlama-1.1B-Chat-v1.0. By default it achieved around 0.20f1-score. Fine-tuned version achieved 0.54 f1-score in less than 4 minutes of fine-tuning on a single A100