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

    Neuronpedia

    INDEX
    Explanations

    listener, you, rogue, node, authors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    il
    0.49
    ou
    0.44
    -
    0.44
    ellt
    0.44
     belly
    0.44
     custodian
    0.43
     retailer
    0.43
     forgo
    0.42
     hasn
    0.41
    ræ
    0.40
    POSITIVE LOGITS
     みんな
    0.52
     Golpe
    0.51
     прозра
    0.49
    腖
    0.49
    㛣
    0.49
     მაშინ
    0.48
    车
    0.47
     বাড়ছে
    0.46
    䏲
    0.46
     Ман
    0.45
    Act Density 0.001%

    No Known Activations