© 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. 10-GEMMASCOPE-RES-131K
    4. 1475
    Prev
    Next
    INDEX
    Explanations

    code author names

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

    The neuron detects Jupyter notebook JSON field names (e.g. “cell_type,” “execution_count,” “source,” 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_10/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
     painstaking
    -1.47
     luxus
    -1.46
     estaban
    -1.46
     comprimidos
    -1.43
     bagus
    -1.41
    Pemer
    -1.39
    ينما
    -1.38
     músculo
    -1.37
    焢
    -1.35
     lentes
    -1.34
    POSITIVE LOGITS
    мым
    1.74
    illées
    1.45
    地在
    1.43
    onaise
    1.41
    ;',
    1.41
    ificato
    1.40
     gần
    1.39
    ;$
    1.37
    ’-
    1.36
    makeatletter
    1.35
    Activations Density 0.003%

    No Known Activations