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

    Neuronpedia

    INDEX
    Explanations

    terms related to reaching agreements or settlements

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "compromise" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ahon
    -0.85
    vals
    -0.74
    asters
    -0.72
    aur
    -0.71
    odes
    -0.71
    umin
    -0.71
    soon
    -0.68
    aver
    -0.67
    uf
    -0.67
    lining
    -0.67
    POSITIVE LOGITS
     compromise
    1.16
     compromises
    1.11
    romising
    0.84
    EngineDebug
    0.82
     compromising
    0.81
     comprom
    0.77
     negotiator
    0.75
    smanship
    0.74
     shenan
    0.73
     negotiators
    0.72
    Act Density 0.009%

    No Known Activations