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

    Neuronpedia

    INDEX
    Explanations

    years'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    quantidade
    0.53
     ediyoruz
    0.49
    tear
    0.46
    inicio
    0.44
    rafos
    0.44
     хотим
    0.43
     треба
    0.43
     queremos
    0.43
     Lembrando
    0.42
     possíveis
    0.42
    POSITIVE LOGITS
     was
    0.47
    This
    0.44
     Design
    0.43
     Hill
    0.41
    was
    0.40
     Kitchen
    0.40
     This
    0.39
     or
    0.39
     Cox
    0.39
     Cult
    0.39
    Act Density 0.007%

    No Known Activations