github
/posts /about
Posts

    Mixture of Experts for Clowns (at a Circus)

    2023-12-14

    :: #mistral, #mixtral, #crime, #moe

    Reward Models: Alignment for the GPU Poor

    2023-10-16

    :: #alignment, #reward-model

    Chai Reward Model Training

    2023-10-12

    :: #training, #reward-model

    On Frankenllama

    2023-08-28

    :: #llama, #crime