Quantization made by Richard Erkhov.
SKY-Ko-Solar-10.7B-lora - GGUF
Original model description:
license: apache-2.0
language:
Model Overview Model Name: SKY-Ko-Solar-10.7B-lora (Fine-Tuned with LoRA)
Model Type: Language Model
Base Model: Solar-10.7B
Fine-Tuning Technique: Low-Rank Adaptation (LoRA)
Description This model is based on the Solar-10.7B, a state-of-the-art language model, and has been fine-tuned using the LoRA. LoRA allows for efficient fine-tuning with fewer parameters, making the model more adaptable and faster to deploy for specific tasks.
Train GPU: A100
Training Time: 15 hours
