© 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. 34-GEMMASCOPE-RES-131K
    7. 115788
    Prev
    Next
    INDEX
    Explanations

    on, off, onsite, on-topic

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

    The neuron is tuned to spot numeric tokens—especially multi-digit numbers and hyphenated number ranges—in the text.

    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
     löyty
    -0.75
    endidikan
    -0.75
     师
    -0.73
     Simón
    -0.72
     mahdol
    -0.72
    resolver
    -0.71
    ˾
    -0.69
    Envio
    -0.68
     Racine
    -0.68
    licated
    -0.68
    POSITIVE LOGITS
     On
    2.31
    On
    2.14
     ON
    2.11
     on
    1.93
    オン
    1.55
    ON
    1.53
    setOn
    1.48
     onsite
    1.46
    on
    1.45
     ons
    1.42
    Activations Density 0.059%

    No Known Activations