© 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. 22-GEMMASCOPE-RES-131K
    7. 65503
    Prev
    Next
    INDEX
    Explanations

    preparation details or structure

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

    The neuron strongly activates on numeric tokens—digits, numbers, and quantitative measurements.

    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
     and
    -1.22
     without
    -1.14
     from
    -1.14
     wiers
    -1.13
     only
    -1.08
     on
    -1.06
     first
    -1.05
     garantía
    -1.03
    <strong>
    -1.02
     at
    -1.02
    POSITIVE LOGITS
     palav
    1.39
    Preparazione
    1.35
    Однако
    1.26
    Transk
    1.25
    Dettagli
    1.23
    那时候
    1.20
    Alternatively
    1.20
    ┍
    1.20
    Anyone
    1.19
    Approximately
    1.18
    Activations Density 0.009%

    No Known Activations