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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating ongoing actions and project coordination

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Orm
    -0.15
    ajs
    -0.15
    ceed
    -0.14
    acting
    -0.14
    Ïĩι
    -0.14
     Routine
    -0.13
     ÅĤ
    -0.13
    ROUT
    -0.13
     intervene
    -0.13
    escaping
    -0.13
    POSITIVE LOGITS
     working
    0.36
     final
    0.31
     hard
    0.28
     busy
    0.28
    working
    0.27
     Working
    0.27
    Working
    0.27
     work
    0.26
    -working
    0.25
    final
    0.24
    Act Density 0.135%

    No Known Activations