© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32944
    Prev
    Next
    INDEX
    Explanations

    dimensional mathematical descriptions

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

    The neuron fires on tokens that represent numbers or numeric references (e.g. digits, ordinals, decimal values).

    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
     staffel
    -1.19
     terminer
    -1.17
     baden
    -1.16
    nade
    -1.11
     myriad
    -1.10
     unending
    -1.10
     別注
    -1.09
    を使って
    -1.09
     ludicrous
    -1.09
     ridiculously
    -1.07
    POSITIVE LOGITS
    There
    1.38
     in
    1.38
     there
    1.27
     until
    1.21
     with
    1.21
     can
    1.19
    After
    1.16
    While
    1.16
     how
    1.16
     How
    1.15
    Activations Density 0.040%

    No Known Activations