© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 70155
    Prev
    Next
    INDEX
    Explanations

    License, ACTION

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

    The neuron is a detector for numeric tokens—especially multi‐digit numbers such as years, version numbers, or revision identifiers.

    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
    definecolor
    -0.80
    acan
    -0.73
    üenza
    -0.72
    ម
    -0.71
    魔
    -0.71
     macular
    -0.71
    ্ন
    -0.70
     brawl
    -0.69
    mandala
    -0.68
    Natasha
    -0.66
    POSITIVE LOGITS
     Unter
    0.92
    Unter
    0.81
     ped
    0.79
     Patricia
    0.73
    Under
    0.73
    HttpResponse
    0.72
    ped
    0.71
    PathParam
    0.71
     Courtney
    0.69
     luc
    0.66
    Activations Density 0.351%

    No Known Activations