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

    Neuronpedia

    INDEX
    Explanations

    reasons for refusal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ेंगू
    0.42
    以便
    0.40
     éventuellement
    0.40
    收集
    0.38
     magnétique
    0.38
    用
    0.38
     получать
    0.37
    有点
    0.37
     Depending
    0.37
     হালকা
    0.37
    POSITIVE LOGITS
     refusal
    0.96
     rejection
    0.96
     رفض
    0.91
     rejects
    0.89
     reject
    0.86
     rejected
    0.86
     rifi
    0.85
     refused
    0.84
     rejecting
    0.84
     refuses
    0.82
    Act Density 0.645%

    No Known Activations