© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38696
    Prev
    Next
    INDEX
    Explanations

    LEFT, right, 0, 1 sequence

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

    The neuron activates on numeric tokens (both integers and decimals) in code or text, i.e. it detects numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ikasi
    -1.04
    呲
    -1.02
     באתר
    -1.02
    нок
    -0.97
    caster
    -0.95
     даже
    -0.94
    мат
    -0.94
    zeera
    -0.93
    Glob
    -0.92
    ésultat
    -0.92
    POSITIVE LOGITS
     zincir
    0.97
     übrigen
    0.94
     gezwungen
    0.90
     direction
    0.90
     kullanılır
    0.90
     quotidienne
    0.89
    stache
    0.86
     deği
    0.86
     backwards
    0.85
    Kosten
    0.84
    Activations Density 0.002%

    No Known Activations