Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    settings or contexts involving technical specifications or instructions, such as shooting modes, camera details, and manual controls

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    tec
    -0.75
    rued
    -0.75
    bern
    -0.69
    obi
    -0.68
    arta
    -0.65
    cit
    -0.64
     Afric
    -0.63
    ãĥ¼ãĥ³
    -0.63
    iolet
    -0.63
    earth
    -0.63
    POSITIVE LOGITS
     treating
    1.38
     distributing
    1.35
     transferring
    1.35
     creating
    1.34
     owning
    1.33
     importing
    1.33
     stopping
    1.33
     putting
    1.31
     placing
    1.30
     reacting
    1.30
    Act Density 0.336%

    No Known Activations