© 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. 25261
    Prev
    Next
    INDEX
    Explanations

    programming concepts and code structure

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

    This neuron primarily activates on numeric tokens—i.e. literal numbers and digit sequences—throughout code and text.

    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
    -1.47
     all
    -1.33
     started
    -1.31
     first
    -1.31
     after
    -1.13
     some
    -1.13
    ……。
    -1.12
    fokus
    -1.10
     what
    -1.09
     on
    -1.06
    POSITIVE LOGITS
     ridiculously
    1.39
     Dinas
    1.31
     também
    1.28
     delightfully
    1.27
     conduire
    1.26
     deliciously
    1.25
    沵
    1.20
     unbelievably
    1.19
     arrêter
    1.19
     donnent
    1.19
    Activations Density 0.002%

    No Known Activations