© 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. 65208
    Prev
    Next
    INDEX
    Explanations

    customized shower designs

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

    The neuron strongly activates on the phrase “How to,” i.e. the opening of “how to” questions or tutorials.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     cliquer
    -1.44
     agir
    -1.22
    izou
    -1.19
    Ꮴ
    -1.16
     Préférences
    -1.16
     vacunas
    -1.14
     ſou
    -1.14
     réessayer
    -1.13
     imprimé
    -1.13
     perſon
    -1.13
    POSITIVE LOGITS
     only
    1.36
     will
    1.27
     mainly
    1.16
     so
    1.16
     really
    1.13
     primarily
    1.12
     just
    1.11
     in
    1.06
     very
    1.05
     before
    1.05
    Activations Density 0.040%

    No Known Activations