© 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. 14596
    Prev
    Next
    INDEX
    Explanations

    dollar amounts and quantities

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

    The neuron is tuned to detect multi-digit numerical tokens (e.g. statistics, scores, or large numbers).

    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
     one
    -2.36
     only
    -2.28
     two
    -1.95
     first
    -1.91
     when
    -1.62
     which
    -1.61
     new
    -1.50
     less
    -1.46
     ONLY
    -1.44
     used
    -1.43
    POSITIVE LOGITS
     gigantic
    1.46
     boister
    1.38
    öhnen
    1.38
    inerario
    1.36
     enormes
    1.36
    områ
    1.34
    鶫
    1.34
    ENTIALS
    1.33
     ogrom
    1.33
     fabulous
    1.33
    Activations Density 0.090%

    No Known Activations