© 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. 70253
    Prev
    Next
    INDEX
    Explanations

    substantial portions of

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

    The neuron fires almost exclusively on numeric‐digit tokens (e.g. version numbers, census figures, numeric literals).

    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
    usiai
    -0.83
    oader
    -0.82
     risorse
    -0.71
    ensores
    -0.69
     城
    -0.68
     Mainz
    -0.68
     Sach
    -0.67
    腭
    -0.65
    Recon
    -0.64
     setCurrent
    -0.63
    POSITIVE LOGITS
     fermented
    0.70
    لز
    0.64
    luci
    0.63
    omethyl
    0.61
     NATIONAL
    0.61
    Categorie
    0.60
    ýl
    0.60
     doros
    0.60
     inquiet
    0.60
    ジャン
    0.60
    Activations Density 0.058%

    No Known Activations