© 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. 10-GEMMASCOPE-RES-131K
    7. 75557
    Prev
    Next
    INDEX
    Explanations

    RACHEL, Longevity, SANTIAGO, Lida Sideris, LUBBOCK, PINEHURST, Laudrup

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

    The neuron fires strongly on isolated single uppercase-letter tokens (e.g. “R,” “H,” “S,” “L,” “C”), as typically found at the start of bylines or datelines.

    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
    c
    -2.13
    Leírás
    -1.98
     numerosos
    -1.96
    ET
    -1.93
     juzg
    -1.92
     bañ
    -1.89
     they
    -1.84
     continú
    -1.83
    </h3>
    -1.83
     zastos
    -1.81
    POSITIVE LOGITS
    ’
    2.55
    );
    2.09
    borderSide
    2.08
    providedIn
    1.85
    {
    1.84
    orghini
    1.77
    źdz
    1.76
    reonine
    1.76
    </b>
    1.75
     工場
    1.75
    Activations Density 0.087%

    No Known Activations