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

    Neuronpedia

    INDEX
    Explanations

    repetitive use of the word "ever" in various contexts

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

    ever followed by adjective

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mList
    -0.52
    tiac
    -0.49
    PII
    -0.47
    matitis
    -0.47
    wiches
    -0.46
    rull
    -0.46
    doin
    -0.45
     princip
    -0.45
     Jiao
    -0.45
    suke
    -0.45
    POSITIVE LOGITS
     Ever
    0.98
    Ever
    0.97
     EVER
    0.91
     Everett
    0.82
     ever
    0.81
    ever
    0.78
     Evers
    0.74
    EVER
    0.71
     Everest
    0.66
     jamás
    0.62
    Act Density 0.004%

    No Known Activations