llama-Augmentoolkit-MilitaryModel-Demo-NotUndertrained

This model achieves the following results on the evaluation set:

• Loss: 0.6264

This is a less-undertrained version of one of the demo factual models (the military one). Both such models were a bit undertrained. This one suffers from that less and should produce better results (theoretically, I have not tested it yet). Same prompt as the military one. Try this model out!