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

    Neuronpedia

    INDEX
    Explanations

    the slightest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     塩
    0.59
    Enabled
    0.57
     either
    0.54
    avour
    0.53
    த்தது
    0.53
    either
    0.52
    both
    0.52
     Similar
    0.52
     వీ
    0.52
     项目
    0.52
    POSITIVE LOGITS
     slightest
    0.78
     offending
    0.72
     outermost
    0.71
     delicate
    0.69
     intensity
    0.68
     effects
    0.68
     insides
    0.68
     underside
    0.68
     dreaded
    0.67
     outer
    0.67
    Act Density 0.504%

    No Known Activations