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

    Neuronpedia

    INDEX
    Explanations

    do have

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     او
    0.41
     Everyone
    0.40
    大于
    0.39
     everyone
    0.36
     mismos
    0.36
    Streams
    0.36
     whats
    0.35
     anything
    0.35
     sobald
    0.35
     బహి
    0.34
    POSITIVE LOGITS
     unfortunately
    0.42
     however
    0.41
    however
    0.39
     viszont
    0.38
     admitted
    0.38
     však
    0.38
     indeed
    0.38
     cependant
    0.37
     porém
    0.37
     definitely
    0.36
    Act Density 0.055%

    No Known Activations