© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 68975
    Prev
    Next
    INDEX
    Explanations

    mathematical and programming expressions

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

    The neuron fires on punctuation and operator symbols used as separators (e.g. commas, equals signs, arithmetic operators) in mathematical or formula‐like contexts.

    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
    0
    -1.59
    軋
    -1.09
    攢
    -1.05
    一口气
    -1.04
    甕
    -0.99
    は
    -0.98
    を
    -0.98
    给我们
    -0.98
     explosives
    -0.98
    ってしまう
    -0.97
    POSITIVE LOGITS
     The
    1.04
     continúas
    0.94
    ownika
    0.93
    を探す
    0.92
    ごとの
    0.90
     transfé
    0.88
    stě
    0.88
    ⣾
    0.88
     and
    0.88
    ご紹介
    0.86
    Activations Density 0.019%

    No Known Activations