© 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. 43904
    Prev
    Next
    INDEX
    Explanations

    steps taken

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

    This neuron fires on the past-participle “taken.”

    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
    </strong>
    -2.11
    |
    -2.08
    {
    -1.91
    which
    -1.81
    m
    -1.80
    y
    -1.80
    d
    -1.79
    se
    -1.79
    <b>
    -1.74
     robić
    -1.71
    POSITIVE LOGITS
    
    1.75
    vantaged
    1.72
    нений
    1.71
    鲎
    1.71
    пыта
    1.71
    ቍ
    1.69
     própri
    1.68
    ཫ
    1.65
    
    1.63
    驺
    1.62
    Activations Density 0.008%

    No Known Activations