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

    Neuronpedia

    INDEX
    Explanations

    cache context resolve

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    跎
    0.58
     ا
    0.53
     esimerk
    0.48
     ایمان
    0.48
    jasama
    0.47
     harte
    0.47
     njemu
    0.47
    ハム
    0.46
    בּ
    0.46
    醗
    0.46
    POSITIVE LOGITS
    Cache
    0.70
    Context
    0.61
    Resolve
    0.61
    cache
    0.59
    Invalid
    0.58
    Cached
    0.58
     resolved
    0.58
    resolved
    0.57
     cached
    0.57
    context
    0.54
    Act Density 0.208%

    No Known Activations