© 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. 44069
    Prev
    Next
    INDEX
    Explanations

    iPhone 6, 8, Windows, Yosemite

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

    The neuron fires on standalone numeric tokens, especially when they denote model numbers, software versions, or sequel identifiers.

    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
     ehemaligen
    -1.37
     aprile
    -1.31
     vuonna
    -1.29
     the
    -1.26
    abinieri
    -1.24
     iden
    -1.23
     cimetière
    -1.23
     calciatore
    -1.20
     hypocritical
    -1.19
     если
    -1.18
    POSITIVE LOGITS
    のように
    1.43
     than
    1.30
    えず
    1.26
    ている
    1.23
     example
    1.22
     mô
    1.21
     parachoque
    1.20
    逻
    1.18
     well
    1.17
     onCancelled
    1.16
    Activations Density 0.129%

    No Known Activations