© 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. 34345
    Prev
    Next
    INDEX
    Explanations

    building, growing, running, printing

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

    The neuron activates on action‐oriented verbs describing operations or processes (e.g. constructed, grow, opening, run, flow).

    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
    分辨
    -0.92
    瀏
    -0.91
     雞
    -0.90
    tony
    -0.90
     своя
    -0.89
     konsek
    -0.89
    föl
    -0.89
     modrá
    -0.88
     twins
    -0.86
     mög
    -0.85
    POSITIVE LOGITS
     they
    1.30
     these
    1.30
     vibra
    1.13
     они
    1.05
     extraordin
    1.05
     of
    1.04
    ζει
    0.99
     reputa
    0.97
     valo
    0.95
     caro
    0.93
    Activations Density 0.077%

    No Known Activations