© 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. 72443
    Prev
    Next
    INDEX
    Explanations

    code snippets

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

    The neuron is detecting numeric literals (both integers and 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
    蕖
    -2.38
    
    -2.25
     trám
    -2.19
     nowy
    -2.13
     naszych
    -2.09
     naszego
    -2.09
     iw
    -2.08
     utilisé
    -2.06
    泐
    -2.06
     dwa
    -2.05
    POSITIVE LOGITS
     in
    3.42
     at
    2.86
     from
    2.70
     what
    2.66
     after
    2.63
     during
    2.56
     by
    2.56
     all
    2.55
     with
    2.52
     During
    2.39
    Activations Density 0.010%

    No Known Activations