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

    Neuronpedia

    INDEX
    Explanations

    past actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     furent
    0.73
    되었다
    0.72
    其
    0.67
     wurden
    0.64
    されている
    0.64
     되었다
    0.64
     furono
    0.62
     byly
    0.60
    变为
    0.60
     Prove
    0.60
    POSITIVE LOGITS
     tried
    0.74
     watched
    0.68
     known
    0.67
     heard
    0.66
     imagined
    0.65
     researched
    0.63
     talked
    0.63
     admired
    0.62
     considered
    0.62
     thought
    0.61
    Act Density 0.133%

    No Known Activations