© 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. 10-GEMMASCOPE-RES-131K
    7. 85432
    Prev
    Next
    INDEX
    Explanations

    time stamps and identifiers

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

    This neuron responds to numeric characters or digit sequences (e.g. version numbers, IDs, filenames) in technical or code‐like contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     we
    -2.19
     dañ
    -1.96
     T
    -1.90
     M
    -1.85
     having
    -1.80
     had
    -1.80
     del
    -1.79
     N
    -1.78
     our
    -1.78
     is
    -1.77
    POSITIVE LOGITS
     Eſ
    3.00
     Inſ
    2.63
    蝼
    2.52
    ૉ
    2.47
     chré
    2.44
    謚
    2.30
     Conſ
    2.27
    ſo
    2.23
     ſer
    2.23
     笔
    2.16
    Activations Density 0.055%

    No Known Activations