© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129527
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively fires on standalone numeric tokens (multi‐digit numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Heat lightning

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     if
    -1.38
     or
    -1.31
     after
    -1.30
    telep
    -1.17
     when
    -1.16
     what
    -1.09
     which
    -1.08
     to
    -1.06
     there
    -1.04
     this
    -1.02
    POSITIVE LOGITS
     boister
    1.17
     astounding
    1.13
     opulent
    1.09
    rea
    1.08
     همچنین
    1.07
     tumultuous
    1.07
     splendid
    1.04
     ensuing
    1.03
    웡
    1.02
    ri
    1.02
    Activations Density 0.001%

    No Known Activations