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

    Neuronpedia

    INDEX
    Explanations

    firewall named, hand side, stopping attacks, wand, hour

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ются
    0.68
    ется
    0.52
    ewhere
    0.52
    <0x07>
    0.50
    ляць
    0.49
    icut
    0.49
    s
    0.48
    icho
    0.48
    oids
    0.47
    ிற
    0.46
    POSITIVE LOGITS
     পু
    0.53
     ഉള്‍
    0.53
    ↵↵
    0.52
     HRV
    0.51
     ഉൾ
    0.49
     હ
    0.46
     ଶ
    0.45
     স্ব
    0.45
     oui
    0.44
     م
    0.44
    Act Density 0.000%

    No Known Activations