© 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. 92997
    Prev
    Next
    INDEX
    Explanations

    Category classifications

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

    The neuron selectively activates on numeric tokens—especially multi‐digit numbers like years and dates.

    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
     olur
    -0.86
     maksimum
    -0.85
    tifkan
    -0.84
    ölkerung
    -0.82
    laş
    -0.80
    ersonal
    -0.79
     мульт
    -0.79
     están
    -0.78
     intelligente
    -0.78
     ferie
    -0.78
    POSITIVE LOGITS
     bring
    0.79
    聲音
    0.78
    づけ
    0.74
     alongside
    0.73
    ütz
    0.72
     fifth
    0.71
    んです
    0.71
     second
    0.70
    gear
    0.70
     aquilo
    0.70
    Activations Density 0.018%

    No Known Activations