© 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. 66134
    Prev
    Next
    INDEX
    Explanations

    multi band, I received, perfect can

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

    This neuron is highly tuned to spotting the “e.g.” abbreviation (the “g” plus its following period).

    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
     people
    -1.36
     first
    -1.31
    Without
    -1.27
     their
    -1.27
     without
    -1.24
    At
    -1.23
     Without
    -1.20
    Sometimes
    -1.14
    Since
    -1.14
    Despite
    -1.12
    POSITIVE LOGITS
     balık
    1.25
     verschillende
    1.19
    
    1.05
     różnych
    1.04
    侄
    1.03
     prendre
    1.02
     alcuni
    1.00
     boty
    1.00
    DAVID
    0.99
     صفحة
    0.99
    Activations Density 0.000%

    No Known Activations