© 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. 3787
    Prev
    Next
    INDEX
    Explanations

    code, Q:, legal citations, news headlines, medical terms

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

    neural network and convolutional layer terminology, particularly numbers associated with layer identifiers.

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

    The neuron fires on numeric tokens, particularly digits and numbers used in code (e.g., layer indices or dimensions).

    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
    ↵↵↵
    -1.91
     zarówno
    -1.89
     of
    -1.66
    tratamiento
    -1.65
     både
    -1.59
    子は
    -1.59
     gewerb
    -1.55
    (\"
    -1.54
     performs
    -1.54
     plans
    -1.54
    POSITIVE LOGITS
     Bandwidth
    1.40
    鯓
    1.37
    姦
    1.36
    orado
    1.35
     kdo
    1.34
     可爱
    1.32
     ambivalent
    1.32
     You
    1.32
    现在的
    1.31
    ナップ
    1.30
    Activations Density 0.002%

    No Known Activations