© 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. 66977
    Prev
    Next
    INDEX
    Explanations

    code separators and delimiters

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

    The neuron fires on numeric tokens (especially floating‐point or version‐style numbers) in source code.

    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
     who
    -1.98
     then
    -1.63
    latimes
    -1.53
    OrNil
    -1.48
     яка
    -1.46
     but
    -1.43
     EconPapers
    -1.34
    ೢ
    -1.34
    抜粋
    -1.34
    erunner
    -1.31
    POSITIVE LOGITS
     that
    1.59
    姆
    1.36
     tuta
    1.34
     harap
    1.31
     kasar
    1.30
     kolye
    1.26
    ム
    1.26
     daga
    1.20
    Youtube
    1.19
     maksi
    1.19
    Activations Density 0.151%

    No Known Activations