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

    Neuronpedia

    INDEX
    Explanations

    but, however

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Without
    0.49
    Without
    0.49
    without
    0.48
     essentially
    0.47
     ohne
    0.46
     pertanto
    0.46
     Essentially
    0.45
    Rationale
    0.45
     без
    0.44
     ছাড়াই
    0.43
    POSITIVE LOGITS
     Importantly
    0.63
     contrary
    0.62
     importantly
    0.61
     crucially
    0.56
     simply
    0.55
     saying
    0.54
     단순히
    0.54
     এখানেই
    0.51
     просто
    0.50
    simply
    0.50
    Act Density 0.059%

    No Known Activations