© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85948
    Prev
    Next
    INDEX
    Explanations

    box or boxes

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

    The neuron is keyed to the literal token “box,” firing strongly whenever it encounters that specific word (often in class or id names).

    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
     leone
    -2.34
     have
    -2.20
    didn
    -2.19
     pegat
    -2.06
    ”“
    -2.05
     lanka
    -2.03
     tzu
    -2.00
     vanden
    -1.94
     lorenzo
    -1.92
    D
    -1.91
    POSITIVE LOGITS
     box
    2.48
     einem
    2.34
     boxes
    2.23
     katkı
    2.19
     otras
    2.19
     {
    2.19
     the
    2.16
    ꔹ
    1.97
     altre
    1.94
     Then
    1.93
    Activations Density 0.028%

    No Known Activations