© 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. 34-GEMMASCOPE-RES-131K
    7. 75959
    Prev
    Next
    INDEX
    Explanations

    names of places and numbers

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

    The neuron is detecting numeric tokens (digits and numbers, such as years, counts, or standalone digits).

    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
    laubnis
    -1.00
     andet
    -1.00
     signify
    -1.00
     نکنید
    -0.95
    隴
    -0.95
    ชอบ
    -0.95
     nhan
    -0.94
    влаж
    -0.94
    🧆
    -0.93
    ộp
    -0.93
    POSITIVE LOGITS
     иллюстрация
    0.91
     Для
    0.90
    <b>
    0.88
     erklär
    0.87
    MENAFN
    0.86
     Songtext
    0.85
     косме
    0.85
     Можно
    0.85
     Однако
    0.85
    Ĭ
    0.85
    Activations Density 0.034%

    No Known Activations