© 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. 38372
    Prev
    Next
    INDEX
    Explanations

    numbers and special characters

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

    The neuron flags tokens belonging to technical metadata and code snippets—e.g. license names, URLs, file‐path markers, config keys, and other code/documentation artifacts.

    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
    addTo
    -0.86
     Беларуси
    -0.75
    向
    -0.71
    文章
    -0.68
    물
    -0.68
    Aaron
    -0.67
     向
    -0.66
    <0xE2>
    -0.66
     /^[
    -0.65
     onPage
    -0.65
    POSITIVE LOGITS
    磚
    0.78
    曾
    0.73
     Verpflich
    0.73
     جمعیت
    0.70
    芩
    0.69
    ực
    0.69
     Crime
    0.67
     twentieth
    0.66
    ギ
    0.65
     Cronin
    0.65
    Activations Density 0.526%

    No Known Activations