© 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. 22-GEMMASCOPE-RES-131K
    4. 39360
    Prev
    Next
    INDEX
    Explanations

    foreign terms and punctuation

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

    The neuron fires on numerical tokens (dates, figures, percentages, monetary amounts, etc.).

    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
     people
    -1.14
     your
    -1.11
     because
    -1.02
    ก็
    -0.97
    をご覧ください
    -0.96
     When
    -0.94
     gwaran
    -0.93
     vivía
    -0.93
    餐廳
    -0.92
     соціаль
    -0.91
    POSITIVE LOGITS
    ливо
    1.09
     dimensión
    1.07
     quidem
    1.03
    ofag
    1.03
     größe
    1.03
     دریافت‌شده
    1.02
     philippe
    1.02
     verde
    1.01
    MQTT
    1.01
    frica
    1.00
    Activations Density 0.026%

    No Known Activations