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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a need or requirement for action or resolution

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating necessity or unresolved issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gdala
    -0.86
    deen
    -0.76
    azar
    -0.71
    SHIP
    -0.68
    ulty
    -0.67
    owship
    -0.62
    ILE
    -0.62
    ccording
    -0.62
    theless
    -0.61
    RANT
    -0.61
    POSITIVE LOGITS
    lessly
    1.06
     updating
    1.03
     tweaking
    1.02
     fixing
    0.97
     correcting
    0.96
     clarification
    0.93
     refinement
    0.92
     adjustment
    0.91
     adjusting
    0.90
     correction
    0.89
    Act Density 0.047%

    No Known Activations