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

    Neuronpedia

    INDEX
    Explanations

    phrases related to expectations and their fulfillment or exceedance

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

    deviations from expectations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    WillAppear
    -0.50
    EndContext
    -0.48
     avoient
    -0.47
     يتيمه
    -0.44
     abstrato
    -0.42
     otomatig
    -0.40
     ſol
    -0.39
     myſelf
    -0.39
    ябре
    -0.38
     verbeteren
    -0.38
    POSITIVE LOGITS
     unexpectedly
    0.70
     unexpected
    0.69
     disappointments
    0.61
     Unexpected
    0.61
    Unexpected
    0.59
    unexpected
    0.58
     unforeseen
    0.55
     surprises
    0.55
     surprise
    0.55
     Überras
    0.54
    Act Density 0.307%

    No Known Activations