© 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. 39411
    Prev
    Next
    INDEX
    Explanations

    São Paulo, carcinomas, mobile

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

    numbers and numerical elements within text.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is primarily triggered by subword tokens that mark the start of words or non-standard characters—i.e. prefixes, digits, punctuation (like apostrophes), and diacritics.

    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
     FOURTH
    -0.91
    初中
    -0.82
    もしれない
    -0.80
     schwarz
    -0.79
     SECONDARY
    -0.77
     Jackpot
    -0.77
    ņa
    -0.75
     SEVENTH
    -0.75
     investissements
    -0.72
     bloss
    -0.72
    POSITIVE LOGITS
     nadzie
    0.75
    IVOS
    0.72
    tangent
    0.71
     Larger
    0.70
    amerikan
    0.69
     २०
    0.69
    technique
    0.69
    íng
    0.69
    pousada
    0.69
     エ
    0.68
    Activations Density 0.096%

    No Known Activations