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

    Neuronpedia

    INDEX
    Explanations

    letting go, release, surrender

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ätze
    0.38
    elige
    0.38
    alam
    0.34
    itei
    0.34
    嵌
    0.34
     appell
    0.34
     redeem
    0.34
    ឌ
    0.34
     tart
    0.33
     جون
    0.33
    POSITIVE LOGITS
     letting
    1.93
     Letting
    1.93
     release
    1.63
     relinquish
    1.60
    release
    1.55
    Release
    1.55
     Release
    1.54
     releasing
    1.52
     surrender
    1.51
     relinqu
    1.51
    Act Density 0.036%

    No Known Activations