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

    Neuronpedia

    INDEX
    Explanations

    expressions related to deficiencies or shortcomings

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

    criticism involving lack or size

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     newBuilder
    -0.40
     mahdol
    -0.33
     also
    -0.33
    RUnlock
    -0.32
     bijzondere
    -0.28
     rechazo
    -0.28
     ibunya
    -0.28
     appui
    -0.27
     erila
    -0.26
     Jerusalén
    -0.26
    POSITIVE LOGITS
    SequentialGroup
    0.74
    adaptiveStyles
    0.66
    tamount
    0.66
    <unused41>
    0.66
    [@BOS@]
    0.65
    <pad>
    0.65
    <unused21>
    0.65
     パンチラ
    0.65
    <unused68>
    0.65
    <unused8>
    0.65
    Act Density 0.014%

    No Known Activations