© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34867
    Prev
    Next
    INDEX
    Explanations

    results and outcomes

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

    The neuron activates on numeric tokens (e.g., scores, statistics, dates, and other number strings).

    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
     new
    -1.20
     you
    -1.16
     now
    -1.15
     still
    -1.11
     проблеми
    -1.03
    {\"
    -1.02
     ingat
    -1.02
     for
    -1.02
     के
    -1.00
     детские
    -1.00
    POSITIVE LOGITS
    alnız
    1.30
     marrom
    1.23
     escuro
    1.20
     vermag
    1.20
     NSCoder
    1.20
     vivere
    1.20
     cinza
    1.20
    ängen
    1.16
    estab
    1.13
     tivi
    1.13
    Activations Density 0.012%

    No Known Activations