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

    Neuronpedia

    INDEX
    Explanations

    job transitions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     измер
    -0.09
     XVI
    -0.08
     Fabric
    -0.08
     Quality
    -0.08
     китай
    -0.08
     бас
    -0.08
     Ris
    -0.07
     karbon
    -0.07
     футб
    -0.07
     Principles
    -0.07
    POSITIVE LOGITS
     remaining
    0.13
     residual
    0.12
     remnants
    0.12
     regained
    0.12
    _remaining
    0.12
    Remaining
    0.12
     revived
    0.12
     salvage
    0.12
     leftover
    0.11
     restarted
    0.11
    Act Density 0.218%

    No Known Activations