© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113641
    Prev
    Next
    INDEX
    Explanations

    obtain a copy

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

    The neuron fires on “obtain … of …” instruction phrases—e.g. “Obtain the first entry of JSON” or “obtain a copy of the License at.”

    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
     resa
    -1.80
     compren
    -1.73
     espace
    -1.65
    these
    -1.63
    ViewById
    -1.60
    臯
    -1.59
     magnifique
    -1.57
     mett
    -1.56
    lowed
    -1.52
    さわ
    -1.49
    POSITIVE LOGITS
     through
    1.90
     Obtaining
    1.85
     from
    1.85
     Obtain
    1.69
     curious
    1.61
     obtain
    1.59
     Actually
    1.48
     Regarding
    1.44
    饧
    1.43
     Pokud
    1.41
    Activations Density 0.131%

    No Known Activations