© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 99104
    Prev
    Next
    INDEX
    Explanations

    Take a

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

    The neuron responds specifically to occurrences of the word “Take” (e.g. as a method name or keyword).

    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
     \
    -3.95
    '
    -3.34
     '
    -3.11
    A
    -2.89
            
    -2.80
    {
    -2.80
     (
    -2.72
     [
    -2.67
    <b>
    -2.61
    と
    -2.58
    POSITIVE LOGITS
    vantaged
    2.77
     夜景
    2.77
     стоковые
    2.64
    浹
    2.52
     unſ
    2.50
    笮
    2.48
    ᯅ
    2.48
    檍
    2.42
    ꦧ
    2.42
     dites
    2.36
    Activations Density 0.013%

    No Known Activations