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

    Neuronpedia

    INDEX
    Explanations

    here are some

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     precisamente
    0.42
     précisément
    0.40
    Analysis
    0.39
    liceerd
    0.39
     Analyse
    0.38
     Analysis
    0.38
     parfaitement
    0.37
     exactamente
    0.36
    iziert
    0.36
    分析
    0.36
    POSITIVE LOGITS
     highlights
    0.91
    highlights
    0.87
     sampling
    0.84
     just
    0.80
     Highlights
    0.78
    sampling
    0.75
    Highlights
    0.75
     partial
    0.75
     parcial
    0.74
     examples
    0.74
    Act Density 0.064%

    No Known Activations