© 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. 73840
    Prev
    Next
    INDEX
    Explanations

    designed by

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

    The neuron fires on boilerplate license/legal text, especially on past‐participle verbs and key terms (“granted,” “drawn,” “License,” “free,” etc.) common in open-source license blocks.

    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
    hubarb
    -0.72
     Hy
    -0.69
    เพ
    -0.68
    umbang
    -0.67
     submitting
    -0.66
    ата
    -0.66
    quec
    -0.65
     Count
    -0.64
     стри
    -0.64
    ără
    -0.63
    POSITIVE LOGITS
    zai
    0.75
    throughout
    0.73
    胪
    0.71
     security
    0.71
     $-.
    0.70
    ⎈
    0.70
     ума
    0.69
    Ϲ
    0.69
    "]="
    0.69
     Loy
    0.68
    Activations Density 0.055%

    No Known Activations