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

    Neuronpedia

    INDEX
    Explanations

    will happen or continue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    for
    0.46
    points
    0.40
    preferences
    0.40
     labels
    0.39
    s
    0.39
    ́s
    0.39
    labels
    0.39
    ภาพ
    0.38
     directives
    0.38
    lines
    0.37
    POSITIVE LOGITS
     survive
    0.62
     be
    0.52
     sobrevivir
    0.51
     blive
    0.49
     succeed
    0.49
     demain
    0.49
     get
    0.47
     bientôt
    0.47
     become
    0.46
    失敗
    0.46
    Act Density 0.118%

    No Known Activations