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

    Neuronpedia

    INDEX
    Explanations

    Chaos

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    -contact
    -0.08
     predictors
    -0.08
    (user
    -0.08
     sie
    -0.07
    	user
    -0.07
     revised
    -0.07
    Overview
    -0.07
     hunn
    -0.07
     overview
    -0.07
     revisar
    -0.07
    POSITIVE LOGITS
     Injection
    0.12
     injection
    0.11
     injected
    0.11
    .Inject
    0.11
     Inject
    0.11
     Fault
    0.10
     perturb
    0.10
     disruptive
    0.10
     disturb
    0.10
    Injection
    0.10
    Act Density 0.002%

    No Known Activations