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

    Neuronpedia

    INDEX
    Explanations

    negative numeric values or accompanying context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    hyphens and CR codes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     indígen
    -0.60
     Infór
    -0.57
     jengibre
    -0.56
     desmotivaciones
    -0.56
     Moscú
    -0.55
     nôtre
    -0.54
     étoient
    -0.54
     plufieurs
    -0.54
     ainfi
    -0.53
     cebolla
    -0.52
    POSITIVE LOGITS
    -
    0.82
    -(
    0.66
    ‐
    0.62
    ()-
    0.61
    -/**
    0.61
    /*-
    0.60
    _
    0.60
    '-
    0.60
     -
    0.59
    ‑
    0.59
    Act Density 0.140%

    No Known Activations