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

    Neuronpedia

    INDEX
    Explanations

    abrupt changes or jerky movements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     εύ
    0.42
     centrally
    0.38
    oodoo
    0.37
    ουλ
    0.37
     മഹ
    0.36
    localScale
    0.36
    োয়
    0.36
    ី
    0.36
    ósz
    0.36
     புராண
    0.36
    POSITIVE LOGITS
     abrupt
    2.11
     jerk
    1.90
     jerky
    1.90
     jer
    1.76
     abruptly
    1.73
     Jer
    1.60
     sudden
    1.55
     sharp
    1.48
    Jer
    1.48
    jer
    1.34
    Act Density 0.052%

    No Known Activations