Hu et al. applied LoRA to the GPT-3 175B model, which reduced the number of trainable parameters by 10,000 times while maintaining inference speed.

Authors

• Person: Not available Organization: Nature
  The construction and refined extraction techniques of knowledge ...

Sources

• The construction and refined extraction techniques of knowledge ... www.nature.com Nature via serper

Referenced by nodes (1)

• GPT-3 concept