© 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. 10-GEMMASCOPE-RES-131K
    7. 103481
    Prev
    Next
    INDEX
    Explanations

    sequences of letters and numbers

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

    lowercase letters that serve as labels or identifiers in technical and scientific text, particularly when following numbers in lists, chemical formulas, figure references, or component designations.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on isolated single‐letter labels (e.g. “A,” “a,” “b,” etc.) used as enumeration or figure/compound identifiers in technical documents.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     to
    -2.52
     as
    -2.25
     not
    -2.06
     where
    -1.90
     in
    -1.81
     such
    -1.80
     from
    -1.79
     will
    -1.76
     has
    -1.74
     became
    -1.73
    POSITIVE LOGITS
    ↵
    2.02
     größe
    1.80
    
    1.66
     $
    1.58
    ఁ
    1.57
     pousse
    1.54
    Fifteen
    1.54
    فوائد
    1.54
    Sabes
    1.52
     "\
    1.52
    Activations Density 0.003%

    No Known Activations