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

    Neuronpedia

    INDEX
    Explanations

    how many, long, why, what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ekki
    0.41
     নানান
    0.39
    விலான
    0.39
     केलेल्या
    0.38
     จน
    0.37
     systému
    0.36
    မဲ့
    0.36
     ikke
    0.36
     لكن
    0.36
     kullanımı
    0.36
    POSITIVE LOGITS
     is
    0.71
     did
    0.69
     are
    0.62
    ?
    0.62
     does
    0.61
     was
    0.60
     would
    0.57
    did
    0.57
     do
    0.51
     were
    0.49
    Act Density 0.104%

    No Known Activations