© 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. 22-GEMMASCOPE-RES-131K
    7. 26536
    Prev
    Next
    INDEX
    Explanations

    algorithmic approach or With the

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

    The neuron fires on tokens from a soccer match preview—especially a Premier League fixture preview.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     första
    -1.20
     other
    -1.17
     this
    -1.17
    🪢
    -1.13
     maybe
    -1.13
     більш
    -1.11
     so
    -1.10
     not
    -1.09
     such
    -1.06
    ähän
    -1.05
    POSITIVE LOGITS
    ſche
    1.16
     dernières
    1.06
     เว
    1.05
     gawas
    1.01
     União
    0.99
    ђе
    0.98
    举报
    0.98
    constexpr
    0.98
     SILVA
    0.96
    outSlope
    0.94
    Activations Density 0.001%

    No Known Activations