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

    Neuronpedia

    INDEX
    Explanations

    allow, skip, expect, describe, each

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     slučaju
    0.39
    Scenario
    0.38
    case
    0.38
    BUL
    0.38
     Illegal
    0.38
     case
    0.38
    illegal
    0.38
    ZIONE
    0.37
    Logic
    0.37
     reactive
    0.36
    POSITIVE LOGITS
    allow
    0.48
    Allow
    0.48
     skip
    0.47
    skip
    0.46
     allow
    0.45
     Allow
    0.45
     позволяют
    0.45
     skips
    0.43
     skipping
    0.42
     vois
    0.41
    Act Density 0.004%

    No Known Activations