© 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. 25728
    Prev
    Next
    INDEX
    Explanations

    loop and variable indices

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

    The neuron is keyed to programming identifiers—especially variable names (e.g. “item”, “i”, “y”, “bytes”, etc.)—highlighting occurrences of variables in 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_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
     of
    -1.43
    BrowserModule
    -1.10
    емо
    -1.05
     made
    -1.05
     Also
    -1.01
    Also
    -1.01
    派的
    -0.93
    Răspunde
    -0.90
     компре
    -0.90
     where
    -0.88
    POSITIVE LOGITS
     ROCKS
    1.06
     Exquisite
    1.02
     rând
    1.02
     Gustave
    1.01
    钌
    0.99
     eventi
    0.98
    そして
    0.98
     reparti
    0.97
    celand
    0.97
    gencias
    0.96
    Activations Density 0.005%

    No Known Activations