© 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. 22-GEMMASCOPE-RES-131K
    7. 54113
    Prev
    Next
    INDEX
    Explanations

    price, work, gay, home, gang, shaped, pressing

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

    The neuron strongly activates on tokens naming or describing extremist ideology or hate‐group labels (e.g. “neo‐Nazi,” “white supremacist”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    呆
    -0.93
    ferous
    -0.93
     assez
    -0.92
    át
    -0.88
    вен
    -0.86
    叭
    -0.85
    own
    -0.84
    ebaran
    -0.83
    ที่มา
    -0.83
     और
    -0.82
    POSITIVE LOGITS
     peinado
    1.05
     vuonna
    1.01
    ükü
    0.96
    imaginary
    0.96
    esas
    0.92
     salvajes
    0.91
    piv
    0.90
    Imaginary
    0.89
    ™,
    0.89
     کے
    0.89
    Activations Density 0.117%

    No Known Activations