Ported the code I wrote using Stable LM 3B to Dolly 2 3b. It worked well on RTX3090 with 64GB of RAM, but exited with out of memory with 16GB of RAM and T4 GPU. #LLM #GPGPU #GPU #coding