© 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. 129555
    Prev
    Next
    INDEX
    Explanations

    based on

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

    The neuron chiefly activates on numeric tokens and quantitative expressions—particularly numbers (including decimals) and related measurement phrases.

    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
     dedica
    -1.59
     durante
    -1.49
     canais
    -1.48
     uniforme
    -1.48
     إذا
    -1.45
    
    -1.45
     revela
    -1.43
    
    -1.43
    Markt
    -1.43
     soooo
    -1.41
    POSITIVE LOGITS
     from
    1.57
     Öffnen
    1.48
    '
    1.47
    (
    1.46
    With
    1.45
    Not
    1.45
     übrigen
    1.36
     only
    1.34
    Also
    1.33
    .
    1.32
    Activations Density 0.033%

    No Known Activations