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

    Neuronpedia

    INDEX
    Explanations

    intent classification and context extraction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ংকর
    0.76
     Masonic
    0.72
     ironically
    0.71
     beaucoup
    0.70
    chner
    0.69
     Ironically
    0.69
     –
    0.66
     gráficos
    0.64
     poquito
    0.64
    十五章
    0.64
    POSITIVE LOGITS
     utterance
    1.22
     sentimento
    1.14
     query
    1.10
     assistant
    1.09
    \'{
    1.08
     evaluation
    1.07
     response
    1.04
     utterances
    1.04
     sentiment
    1.04
     Query
    1.02
    Act Density 2.008%

    No Known Activations