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

    Neuronpedia

    INDEX
    Explanations

    specific subject and its state

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     feeling
    0.64
     curious
    0.57
     impactful
    0.57
     bringing
    0.57
     optimistic
    0.54
     wondering
    0.53
     tweak
    0.52
     picking
    0.51
     convince
    0.51
     factoring
    0.50
    POSITIVE LOGITS
    は
    0.88
    是
    0.76
     は
    0.75
    는
    0.69
     is
    0.69
    은
    0.66
     was
    0.65
     may
    0.64
     wird
    0.64
     must
    0.63
    Act Density 0.320%

    No Known Activations