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

    Neuronpedia

    INDEX
    Explanations

    to build

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    encour
    0.44
    plotting
    0.42
    provides
    0.42
     advisory
    0.41
    seeking
    0.41
    providing
    0.41
     எவ்வாறு
    0.40
     Seeking
    0.39
    delim
    0.39
    ภ
    0.39
    POSITIVE LOGITS
     go
    0.69
     get
    0.64
     REALLY
    0.62
     really
    0.53
     gotten
    0.52
     messed
    0.50
     talk
    0.49
     suka
    0.49
     talked
    0.48
     eat
    0.47
    Act Density 0.047%

    No Known Activations