© 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. 11
    Prev
    Next
    INDEX
    Explanations

    score or ranking

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

    The neuron activates on numeric tokens—especially multi‐digit numbers (scores, stats, dates, etc.).

    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
     resalt
    -0.85
    EFIAPI
    -0.85
    vello
    -0.85
    ন্ত
    -0.84
     potenci
    -0.84
     chaque
    -0.84
    の葉
    -0.83
    にほん
    -0.83
     spad
    -0.83
     botões
    -0.82
    POSITIVE LOGITS
     with
    0.94
    IAS
    0.83
     by
    0.83
    oter
    0.81
     設備
    0.80
    oben
    0.79
    ْم
    0.79
     but
    0.79
    Rgds
    0.77
     Pergamon
    0.77
    Activations Density 0.008%

    No Known Activations