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

    Neuronpedia

    INDEX
    Explanations

    rejection and repelling advances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ัก
    0.60
    রা
    0.57
    ोत्
    0.55
    ू
    0.53
    बलेट
    0.52
    袢
    0.52
     hound
    0.52
     아니
    0.51
     बनवा
    0.51
    明治
    0.49
    POSITIVE LOGITS
     Rejected
    1.04
     rejected
    0.96
     rejection
    0.86
    rejected
    0.86
    Rejected
    0.82
     reject
    0.79
    reject
    0.79
     rejects
    0.78
     Reject
    0.78
     rejet
    0.72
    Act Density 0.027%

    No Known Activations