© 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. 22-GEMMASCOPE-RES-131K
    7. 129049
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on discourse markers that flag inclusion in a broader program or initiative—especially the phrase “part of…” and related terms like “an effort.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    part of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     takiej
    -1.23
    grees
    -1.23
    olding
    -1.22
     RATIO
    -1.18
    会出现
    -1.13
    Télécharger
    -1.11
    habitants
    -1.10
     難
    -1.10
     acestui
    -1.09
    Produk
    -1.08
    POSITIVE LOGITS
     ongoing
    1.74
     a
    1.60
     "
    1.48
     efforts
    1.36
     part
    1.27
     an
    1.19
     /
    1.19
     to
    1.19
     continuing
    1.17
     insgesamt
    1.17
    Activations Density 0.032%

    No Known Activations