© 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. 66761
    Prev
    Next
    INDEX
    Explanations

    foreign words and court documents

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

    The neuron fires strongly on multi‐digit numeric tokens (long sequences of digits, like years or docket 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
     by
    -1.56
     in
    -1.43
     all
    -1.38
     on
    -1.23
     because
    -1.20
     any
    -1.17
     one
    -1.15
     can
    -1.15
     from
    -1.14
     every
    -1.06
    POSITIVE LOGITS
    為に
    1.29
    すっ
    1.20
    튿
    1.10
                
    1.09
     otwor
    1.08
     tôt
    1.06
     trám
    1.05
     uitzicht
    1.05
    凄い
    1.03
     жк
    1.03
    Activations Density 0.019%

    No Known Activations