Train merged LoRa on specific dataset to further improve performance on desired task. This fine-tuning process leverages the pre-trained knowledge encoded in the LoRA adapters.
Description
Reduce the multiplier by 0.3
Train merged LoRa on specific dataset to further improve performance on desired task. This fine-tuning process leverages the pre-trained knowledge encoded in the LoRA adapters.
Reduce the multiplier by 0.3