© 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. 10561
    Prev
    Next
    INDEX
    Explanations

    take a look

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

    The neuron strongly activates on the “Take a …” call-to-action phrase, especially the “a” following “Take.”

    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
    b
    -2.13
    s
    -2.11
    д
    -2.06
    <i>
    -2.05
    要
    -2.02
    0
    -1.98
    >
    -1.87
    H
    -1.79
    .”
    -1.77
    "
    -1.77
    POSITIVE LOGITS
    ynchronously
    1.83
    檍
    1.83
    Стоимость
    1.80
    往上
    1.75
     primas
    1.71
    archiwizowane
    1.69
    Sekarang
    1.69
    🧌
    1.66
    瑧
    1.65
    educated
    1.63
    Activations Density 0.008%

    No Known Activations