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

    Neuronpedia

    INDEX
    Explanations

    positive appraisal and understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Agreed
    0.51
     disagreed
    0.50
    好吧
    0.49
     disagree
    0.46
    的态度
    0.45
     calmed
    0.45
     नाराजगी
    0.44
    ok
    0.44
    Alright
    0.44
     agree
    0.44
    POSITIVE LOGITS
     fascinating
    0.82
     интересный
    0.66
     clever
    0.64
     ingenious
    0.61
     interesting
    0.60
     Interesting
    0.59
    Interesting
    0.59
     Nowadays
    0.57
     интерес
    0.55
     marvelous
    0.54
    Act Density 0.005%

    No Known Activations