© 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. 58402
    Prev
    Next
    INDEX
    Explanations

    sports championships and fields

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

    The neuron activates on numeric tokens—scores, records, dates, measurements—essentially any multi-digit number 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
     kecamatan
    -0.77
    kswagen
    -0.74
     there
    -0.73
    importer
    -0.72
    菓
    -0.70
    刊
    -0.70
     Vorteil
    -0.68
     around
    -0.68
    pos
    -0.66
     dopiero
    -0.66
    POSITIVE LOGITS
     antagonism
    0.86
     prius
    0.80
    ęć
    0.77
     сете
    0.77
    DBA
    0.76
    为什么会
    0.76
    rapid
    0.75
    ᵛ
    0.75
    ilustrasi
    0.75
     sora
    0.75
    Activations Density 0.020%

    No Known Activations