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

    Neuronpedia

    INDEX
    Explanations

    traceback reporting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ходите
    0.51
    lensFlare
    0.45
     kit
    0.42
    kit
    0.41
    vues
    0.41
     kits
    0.40
     baby
    0.40
    searchTerm
    0.38
    baby
    0.38
    kits
    0.38
    POSITIVE LOGITS
     traceback
    1.01
    Trace
    0.81
     Trace
    0.79
     trace
    0.78
     tb
    0.76
    trace
    0.75
     TB
    0.73
    TB
    0.70
    tb
    0.68
     exc
    0.64
    Act Density 0.008%

    No Known Activations