© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 112535
    Prev
    Next
    INDEX
    Explanations

    He performing actions

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

    The neuron strongly activates on third-person subject and relative pronouns (e.g. “he,” “who,” “his,” “will”), effectively signaling references to a male individual as the sentence’s subject.

    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
     it
    -1.67
    回目の
    -1.53
    Важно
    -1.52
     Souhaitez
    -1.47
     dîner
    -1.40
     todav
    -1.40
     壓
    -1.38
    﨑
    -1.37
     Temporada
    -1.36
     pensent
    -1.36
    POSITIVE LOGITS
     to
    1.86
     we
    1.80
     our
    1.64
    杁
    1.43
    </th>
    1.41
    .}(
    1.41
     taken
    1.33
     last
    1.29
               
    1.27
     utilizado
    1.26
    Activations Density 0.064%

    No Known Activations