© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 6806
    Prev
    Next
    INDEX
    Explanations

    speed up or speeding up

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

    The neuron strongly fires on the token “up,” especially when it appears as part of phrasal verbs indicating acceleration (e.g. “speeding up,” “speed up”).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    出了一
    -2.22
    ).
    -2.16
    多么
    -2.11
     terminó
    -1.91
    深处
    -1.90
    汩
    -1.88
     entera
    -1.81
    élection
    -1.77
    imantan
    -1.71
     
    -1.70
    POSITIVE LOGITS
    ଢ
    2.34
    我们
    2.28
    饪
    2.28
    ஂ
    2.28
    gå
    2.23
     Some
    2.23
    敺
    2.14
    凩
    2.11
    蛩
    2.11
     royaltyfri
    2.09
    Activations Density 0.003%

    No Known Activations