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

    Neuronpedia

    INDEX
    Explanations

    therefore answer is

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     prot
    0.43
     you
    0.40
     natu
    0.40
     cas
    0.40
     termina
    0.40
     hope
    0.40
     format
    0.39
     their
    0.39
     mars
    0.39
    Hopefully
    0.39
    POSITIVE LOGITS
    يير
    0.45
     choque
    0.42
     envolvendo
    0.39
     millió
    0.39
    轻轻
    0.38
     علاوه
    0.38
    所有人
    0.38
     เสียง
    0.38
    回应
    0.37
     инвали
    0.37
    Act Density 0.033%

    No Known Activations