© 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. 48710
    Prev
    Next
    INDEX
    Explanations

    class, batch, variants, upgrades

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

    The neuron fires on numeric tokens—particularly multi‐digit figures and numeric model or spec identifiers in technical contexts.

    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
    thylamine
    -0.73
     Coronary
    -0.70
     Autos
    -0.68
    爺
    -0.68
     weißen
    -0.67
     AVL
    -0.66
     regulación
    -0.66
     kleineren
    -0.66
     guesthouse
    -0.66
    elementary
    -0.65
    POSITIVE LOGITS
     variant
    1.40
     Block
    1.27
     variants
    1.20
    variant
    1.19
     upgrade
    1.15
     upgrades
    1.14
    Variant
    1.02
    Block
    1.01
    Upgrade
    1.01
     BLOCK
    1.00
    Activations Density 0.033%

    No Known Activations