© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 101840
    Prev
    Next
    INDEX
    Explanations

    handout or workout

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

    The neuron consistently activates on words naming distributed or supplemental materials (e.g. “handout(s)” and “workout”).

    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>
    -2.72
     avec
    -2.52
    f
    -2.50
     WITH
    -2.50
    an
    -2.47
    一些
    -2.42
    之以
    -2.22
    gebras
    -2.20
    1
    -2.16
    ed
    -2.11
    POSITIVE LOGITS
    </h3>
    3.28
    </strong>
    2.97
    葙
    2.53
    后续
    2.48
     snd
    2.45
    ным
    2.38
    佮
    2.38
    菝
    2.38
    灩
    2.31
    狴
    2.28
    Activations Density 0.004%

    No Known Activations