Our initial approach to LLM is the Llama 2 due to its very high performance, its compactness and its ability for modification through Reinforcement Learning
The overall proposed training mechanism looks as follows
LLAMA 2 language model
Our initial approach to LLM is the Llama 2 due to its very high performance, its compactness and its ability for modification through Reinforcement Learning
The overall proposed training mechanism looks as follows