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

    Neuronpedia

    INDEX
    Explanations

    exact replication

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Actual
    0.51
    Actually
    0.50
    実際に
    0.49
     Actual
    0.49
    Exactly
    0.49
     yaklaşık
    0.48
     actually
    0.47
     actual
    0.47
    approximately
    0.46
     एक्चुअली
    0.46
    POSITIVE LOGITS
    itude
    0.60
    itudes
    0.57
     replicas
    0.54
     matches
    0.48
     replica
    0.48
     match
    0.47
    match
    0.45
    匹配
    0.44
     pinpoint
    0.43
     khớp
    0.42
    Act Density 0.018%

    No Known Activations