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

    Neuronpedia

    INDEX
    Explanations

    instances of incorrectness or inaccuracies in statements or data

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

    incorrectness or inappropriateness

    np_acts-logits-general · gemini-2.0-flash

    incorrect, improper, inaccurate, errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    CEP
    -0.43
     liefer
    -0.42
    autorest
    -0.40
    ArrowToggle
    -0.40
    matchCondition
    -0.39
    ConfigureAwait
    -0.39
     dân
    -0.39
     Willis
    -0.38
     bestand
    -0.38
    ollectionView
    -0.38
    POSITIVE LOGITS
    nappropriate
    0.74
     incorrect
    0.67
     inappropriate
    0.66
     wrong
    0.65
     improper
    0.65
     falsche
    0.65
     ErrInvalid
    0.64
     Improper
    0.63
     WRONG
    0.63
    Inappropriate
    0.62
    Act Density 0.526%

    No Known Activations