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

    Neuronpedia

    INDEX
    Explanations

    occurrences and discussions of collapses or failures in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    breakdown or collapse

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Personendaten
    -0.52
     Employ
    -0.46
    깐
    -0.46
    σθαι
    -0.46
     dul
    -0.45
     esternos
    -0.45
     chl
    -0.45
     resid
    -0.45
    umed
    -0.44
     employ
    -0.44
    POSITIVE LOGITS
     collapse
    1.32
     collapsed
    1.21
     collapses
    1.16
    collapse
    1.16
     collapsing
    1.08
     Collapse
    1.06
    Collapse
    0.87
    collapsed
    0.78
    lapsing
    0.77
    Collapsed
    0.76
    Act Density 0.022%

    No Known Activations