© 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. 10-GEMMASCOPE-RES-131K
    7. 11126
    Prev
    Next
    INDEX
    Explanations

    voxel levels

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

    The neuron activates on any token containing the substring “vox,” especially occurrences of “voxel” or “voxels.”

    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
    豨
    -3.00
    ⾳
    -2.77
     are
    -2.77
    轳
    -2.73
    骜
    -2.59
    炁
    -2.58
    which
    -2.58
    霂
    -2.50
    神
    -2.50
    
    -2.48
    POSITIVE LOGITS
    ⦊
    3.13
    {
    2.92
     
    2.78
    are
    2.52
    )
    2.47
    ’
    2.47
    $
    2.45
     είναι
    2.33
     It
    2.33
    “
    2.27
    Activations Density 0.002%

    No Known Activations