© 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. 15247
    Prev
    Next
    INDEX
    Explanations

    activities and achievements

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

    This neuron fires most strongly on multi‐word gaming-style action phrases—especially ongoing “grinding” descriptions like “running in circles around … looking for rare mobs.”

    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
     you
    -1.94
     then
    -1.88
     immediately
    -1.59
     other
    -1.55
     your
    -1.47
     one
    -1.45
     first
    -1.45
     some
    -1.38
     begy
    -1.34
     Entire
    -1.30
    POSITIVE LOGITS
    綃
    1.23
    ߞ
    1.23
     abonnement
    1.22
    颪
    1.20
     becauſe
    1.20
     offenbar
    1.18
     minuta
    1.17
    laceae
    1.17
    籀
    1.17
     MAIS
    1.15
    Activations Density 0.134%

    No Known Activations