© 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. 34-GEMMASCOPE-RES-131K
    7. 107068
    Prev
    Next
    INDEX
    Explanations

    code and legal documents

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

    The neuron strongly fires on tokens that consist of or include digits—i.e. numeric tokens (dates, counts, IDs, code-line numbers).

    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
    ardonnay
    -0.81
     Sheets
    -0.77
    rtner
    -0.75
    amber
    -0.74
    asier
    -0.73
     Вла
    -0.71
    oq
    -0.70
    pch
    -0.70
     arbets
    -0.70
     matplotlib
    -0.69
    POSITIVE LOGITS
    LEVEL
    0.67
    itable
    0.66
    GF
    0.65
    킵
    0.64
     ggf
    0.63
    mentation
    0.63
    取得
    0.62
    eltjes
    0.62
    sities
    0.60
    SIGNAL
    0.60
    Activations Density 0.057%

    No Known Activations