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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "when" or phrases indicating timing or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Run
    -0.15
    á»ĩu
    -0.14
    Block
    -0.14
     Wizard
    -0.14
    agic
    -0.14
    estic
    -0.14
     Faces
    -0.14
     Duch
    -0.13
     Run
    -0.13
     suff
    -0.13
    POSITIVE LOGITS
    éĮĦ
    0.15
    å½¢
    0.15
    aternion
    0.15
    .scalablytyped
    0.15
    erno
    0.14
    imar
    0.14
    òi
    0.14
    å½ķ
    0.14
    nore
    0.14
    icros
    0.14
    Act Density 0.001%

    No Known Activations