© 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. 40049
    Prev
    Next
    INDEX
    Explanations

    born before dates or places

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

    The neuron detects biographical or résumé-style metadata in a text (names, dates, roles, places of birth, education, positions).

    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
    Our
    -1.52
     dégust
    -1.41
     Проци
    -1.40
     rajut
    -1.39
     our
    -1.34
     etek
    -1.33
    Приятного
    -1.33
    детский
    -1.30
    Моя
    -1.30
    девушка
    -1.29
    POSITIVE LOGITS
    tallet
    1.41
     proteína
    1.38
    BACA
    1.37
    usions
    1.35
     o
    1.34
    ÈNE
    1.27
     amorosa
    1.26
    ּ
    1.23
     vitamina
    1.23
     admire
    1.23
    Activations Density 0.051%

    No Known Activations