© 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. 99820
    Prev
    Next
    INDEX
    Explanations

    actions completed or made possible

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

    The neuron detects mentions of experimental or procedural actions (e.g. “conducted,” “performed,” “achieved”) typically found in scientific study or methods descriptions.

    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
     not
    -1.30
     doesn
    -1.19
     utilized
    -1.17
     perhaps
    -1.14
     supposed
    -1.13
     here
    -1.13
     overwhelmingly
    -1.12
     completamente
    -1.12
     probably
    -1.10
     sistemas
    -1.10
    POSITIVE LOGITS
    ă
    1.10
     ratification
    1.09
    ิต
    1.09
    𝘮
    1.08
     zagran
    1.08
    gaben
    1.08
     gaping
    1.07
    ो
    1.06
     mended
    1.06
     colours
    1.06
    Activations Density 0.377%

    No Known Activations