© 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. 22-GEMMASCOPE-RES-131K
    7. 67452
    Prev
    Next
    INDEX
    Explanations

    created on date

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

    The neuron activates on numeric metadata in file headers—especially creation dates or version numbers (e.g. “Created on Nov 18, 2015”).

    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
     różne
    -0.92
     gratuites
    -0.89
     toalla
    -0.88
    íteni
    -0.88
     gratuits
    -0.86
    jectile
    -0.86
     Fieber
    -0.85
     известных
    -0.85
    ⠀⠀⠀⠀⠀⠀⠀⠀
    -0.85
     this
    -0.85
    POSITIVE LOGITS
    但在
    0.92
     tchèque
    0.91
    数百
    0.91
    瘠
    0.90
     until
    0.89
     had
    0.88
    bruch
    0.88
    停留
    0.86
     and
    0.85
    isier
    0.84
    Activations Density 0.011%

    No Known Activations