© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 54281
    Prev
    Next
    INDEX
    Explanations

    qualities and states

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

    The neuron fires on tokens that are common English derivational suffixes (–ing, –ly, –ful, –ance, –ity, etc.), effectively spotting words formed with those endings.

    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
     rédu
    -1.00
    nope
    -0.96
    ₱
    -0.96
    minha
    -0.95
     dringend
    -0.93
    ান
    -0.93
     corações
    -0.91
     negras
    -0.90
    ganz
    -0.88
    ต่าง
    -0.87
    POSITIVE LOGITS
     and
    1.95
    publicain
    0.87
    arian
    0.85
     connessione
    0.84
     Eſ
    0.84
     during
    0.83
    AUTHORIZED
    0.82
    などを
    0.82
     ifølge
    0.82
     wonderland
    0.82
    Activations Density 0.086%

    No Known Activations