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

    Neuronpedia

    INDEX
    Explanations

    CUDA_VISIBLE_DEVICES

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Wharf
    0.39
     люб
    0.38
     बल
    0.36
     Bradshaw
    0.36
    க்
    0.35
     Lith
    0.35
     خواہ
    0.35
    晾
    0.35
    lith
    0.35
     unanticipated
    0.35
    POSITIVE LOGITS
     Emission
    0.44
    Hell
    0.42
    Emission
    0.41
     Dispersion
    0.40
     emission
    0.39
    hell
    0.39
     Elemente
    0.38
     Neutral
    0.38
     hell
    0.38
    neutral
    0.38
    Act Density 0.001%

    No Known Activations