© 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. 44626
    Prev
    Next
    INDEX
    Explanations

    event times

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

    The neuron fires on standalone numeric tokens (e.g. dates, times, street or house 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_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
     esigen
    -1.60
     mentre
    -1.55
    何时
    -1.50
     žena
    -1.48
     desactivar
    -1.45
     colorear
    -1.45
    éton
    -1.42
     puteți
    -1.36
    sometimes
    -1.34
     există
    -1.34
    POSITIVE LOGITS
    providedIn
    1.48
    ?).
    1.46
     „
    1.40
    ,'"
    1.38
    getParameters
    1.35
    ...,
    1.35
     voulait
    1.30
    菈
    1.30
     forKey
    1.30
     alle
    1.27
    Activations Density 0.050%

    No Known Activations