© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32
    Prev
    Next
    INDEX
    Explanations

    content warnings and explicit labels

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly responds to content‐warning and rating phrases (e.g. “Explicit Content,” “WARNING:,” “Parental guidance,” “adult content advisory”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    brei
    -0.75
    RelativeLayout
    -0.71
    isSelected
    -0.70
     thermometers
    -0.69
    tomy
    -0.69
    zende
    -0.69
     RMSE
    -0.68
    })^{
    -0.67
     jij
    -0.67
    tisch
    -0.66
    POSITIVE LOGITS
     content
    2.38
     warning
    2.28
     mature
    2.27
    Warning
    2.25
    warning
    2.08
     Warning
    2.08
     warnings
    1.99
     contains
    1.98
    WARNING
    1.97
     WARNING
    1.96
    Activations Density 0.085%

    No Known Activations