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

    Neuronpedia

    INDEX
    Explanations

    intents or discussion points

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.54
    {
    0.51
    \
    0.48
     Volume
    0.45
     Batch
    0.45
     Diffusion
    0.44
    $
    0.43
     Verkehr
    0.43
     Breakdown
    0.43
    Volume
    0.42
    POSITIVE LOGITS
     спу
    0.68
    ajem
    0.53
     організа
    0.51
     приема
    0.51
     אמ
    0.50
    нима
    0.50
    ämän
    0.50
    𒌨
    0.50
     словам
    0.49
     마음
    0.49
    Act Density 0.000%

    No Known Activations