© 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. 25871
    Prev
    Next
    INDEX
    Explanations

    closing punctuation marks

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

    This neuron fires specifically on numeric tokens (i.e. number literals) in code snippets.

    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
     from
    -1.68
     begin
    -1.59
     get
    -1.45
     only
    -1.45
     all
    -1.45
     possibly
    -1.45
     to
    -1.40
     while
    -1.36
     like
    -1.34
     obtain
    -1.34
    POSITIVE LOGITS
     ì
    1.34
    久しぶりの
    1.31
    キラキラ
    1.29
    ふと
    1.27
     ۴
    1.27
    recently
    1.23
    さぁ
    1.23
     vack
    1.23
     anunció
    1.23
    appropri
    1.22
    Activations Density 0.011%

    No Known Activations