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

    Neuronpedia

    INDEX
    Explanations

    instead of or instead

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ൊന്നും
    0.45
     Irak
    0.43
     relevancia
    0.41
     kinases
    0.40
     grumpy
    0.40
     fontos
    0.40
     Relevance
    0.40
     különböző
    0.40
     importantes
    0.40
     relev
    0.39
    POSITIVE LOGITS
     вместо
    0.59
    with
    0.56
    を
    0.56
     with
    0.55
     +
    0.51
    Instead
    0.50
    을
    0.50
    instead
    0.49
     instead
    0.49
    으로
    0.48
    Act Density 0.052%

    No Known Activations