© 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. 65834
    Prev
    Next
    INDEX
    Explanations

    declaration or initialization

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

    The neuron is detecting numeric literal tokens (e.g. digits used as constants or array sizes) in code.

    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
     тяжело
    -1.04
     consid
    -1.00
    ੋਂ
    -0.97
     effe
    -0.96
     KOR
    -0.95
     Lieut
    -0.94
     excep
    -0.92
    oarece
    -0.92
     prefent
    -0.90
     considere
    -0.90
    POSITIVE LOGITS
     from
    1.29
     new
    1.20
     many
    1.19
     another
    1.13
     one
    1.13
     gyven
    1.10
     our
    1.09
     first
    1.06
     mūsų
    1.05
     =
    1.05
    Activations Density 0.078%

    No Known Activations