LLM limitations #14
Description
The current EleutherAI/gpt-neo-125M model is offering some limitations, in fact it is not yet possible to reach a fully potential from generative chat. In this context, a new testing/dev environment should be anticipated by improving calculation capacity and potentially GPU specifications.
specs -> RB+eGPU or stronger configuration