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

    Neuronpedia

    INDEX
    Explanations

    cleaning and standardizing text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     marcado
    0.47
     marka
    0.44
     marquée
    0.42
     Mk
    0.41
     marcada
    0.39
     ayam
    0.38
     zeer
    0.38
     ay
    0.38
     postoji
    0.38
     mk
    0.37
    POSITIVE LOGITS
    trim
    0.57
    trimmed
    0.52
    Trim
    0.51
     trimming
    0.50
     eliminate
    0.49
    remove
    0.48
     trimmed
    0.48
     trim
    0.47
    filter
    0.46
     Trim
    0.45
    Act Density 0.362%

    No Known Activations