© 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. 34-GEMMASCOPE-RES-131K
    7. 57929
    Prev
    Next
    INDEX
    Explanations

    scientific and technical articles

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

    academic and scientific publication references, particularly journal citations.

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

    The neuron strongly activates on numeric tokens (multi‐digit numbers, years, figure references, citation numbers, measurements).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     while
    -1.07
     While
    -0.97
    While
    -0.96
     mientras
    -0.86
    linde
    -0.84
    primaryStage
    -0.80
     Sementara
    -0.80
     Während
    -0.79
    レコード
    -0.76
     remain
    -0.73
    POSITIVE LOGITS
     petroleum
    0.94
     przewod
    0.88
     гео
    0.87
     hydrocarbons
    0.82
     Refining
    0.76
     paja
    0.76
    guided
    0.75
     trái
    0.73
     специаль
    0.73
     special
    0.73
    Activations Density 0.005%

    No Known Activations