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

    Neuronpedia

    INDEX
    Explanations

    stroke, collapse, collision

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    に取り組
    0.94
    日子
    0.90
    に進
    0.86
     dexterity
    0.85
    ০০
    0.82
     audacity
    0.80
     fundament
    0.80
    ならば
    0.80
     букмекер
    0.79
    是被
    0.79
    POSITIVE LOGITS
    hanging
    0.87
                        
    0.87
     vadati
    0.86
    apparent
    0.84
    kého
    0.82
    tweets
    0.80
    ajjati
    0.80
    extremely
    0.80
    ik
    0.79
    內的
    0.79
    Act Density 0.330%

    No Known Activations