© 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. 48186
    Prev
    Next
    INDEX
    Explanations

    $. separators

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

    The neuron activates on numeric tokens and related symbols (digits, minus‐signs, primes) in math expressions.

    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
    <td>
    -3.50
     this
    -2.48
     during
    -2.39
     that
    -2.34
     focusing
    -2.33
     because
    -2.25
     their
    -2.23
     these
    -2.19
     ALWAYS
    -2.16
     makes
    -2.14
    POSITIVE LOGITS
    鸲
    2.52
     trám
    2.52
    艄
    2.36
    鲞
    2.33
     queridos
    2.33
    鲩
    2.23
    眛
    2.23
    螃
    2.23
     vivent
    2.22
     contiennent
    2.22
    Activations Density 0.002%

    No Known Activations