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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128932
    Prev
    Next
    INDEX
    Explanations

    The neuron predominantly lights up on numeric literal tokens (i.e. numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    log message

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     comportamento
    -1.05
    sivamente
    -0.98
    ícias
    -0.92
    嘢
    -0.91
    meisters
    -0.89
    bs
    -0.88
    PAH
    -0.87
    stest
    -0.87
    ErrMsg
    -0.85
    ally
    -0.85
    POSITIVE LOGITS
     says
    1.20
    となりました
    1.19
     pity
    1.13
     vermelhas
    1.13
     on
    1.12
     said
    1.10
     süß
    1.10
     brancas
    1.09
    jauh
    1.09
     say
    1.05
    Activations Density 0.005%

    No Known Activations