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

    Neuronpedia

    INDEX
    Explanations

    companion app, ads, diagnostic, computer, book

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     accompanies
    0.81
     accompany
    0.73
     accompaniment
    0.66
     accompanying
    0.65
     сопрово
    0.65
     acompan
    0.64
     accompan
    0.64
     acompañ
    0.63
     accompagne
    0.63
     accompanied
    0.62
    POSITIVE LOGITS
    hong
    0.44
    belle
    0.43
    Tela
    0.43
    पछि
    0.42
    हेही
    0.41
    analisi
    0.41
     बिज
    0.40
    že
    0.40
    Já
    0.40
    व्
    0.40
    Act Density 0.001%

    No Known Activations