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

    Neuronpedia

    INDEX
    Explanations

    obtaining, analyzing, evaluating, creating

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Departure
    0.50
    cvec
    0.46
    departure
    0.45
     multiplicação
    0.42
    xtime
    0.42
    acije
    0.41
    lighthouse
    0.40
     taxed
    0.40
    Petra
    0.40
    ligence
    0.40
    POSITIVE LOGITS
     obtaining
    0.97
     constructing
    0.97
     Obtaining
    0.95
    constructing
    0.83
     Construct
    0.77
     obtain
    0.76
     Obtain
    0.75
    Construct
    0.75
     construct
    0.74
     Analyzing
    0.73
    Act Density 0.001%

    No Known Activations