© 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. 10-GEMMASCOPE-RES-131K
    7. 9668
    Prev
    Next
    INDEX
    Explanations

    a/an noun

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

    This neuron responds to numeric expressions, particularly decimal or floating-point 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_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
    猡
    -2.30
    ⬮
    -1.89
     pérd
    -1.77
    
    -1.69
     情侣
    -1.69
     etek
    -1.68
    ︱
    -1.67
    喁
    -1.65
    ván
    -1.65
     kazak
    -1.63
    POSITIVE LOGITS
     of
    2.88
     .
    1.71
    as
    1.68
     nadzieję
    1.68
    ds
    1.63
     he
    1.50
     past
    1.50
    ことから
    1.49
     hint
    1.48
    0
    1.47
    Activations Density 0.009%

    No Known Activations