search

chaoyi-wu / PMC-LLaMA

The official codes for "PMC-LLaMA: Towards Building Open-source Language Models for Medicine"
549 stars 52 forks source link

Pretraining clarification #28

Open vikigenius opened 5 months ago

vikigenius commented 5 months ago

The paper claims that you trained for 5 epochs on 32 A100 GPUs. Do you have an estimate of how much time it took? And also was it a 40GB A100 or 80 GB?

chaoyi-wu commented 3 months ago

80 GB is used. Around one week.