© 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. 26698
    Prev
    Next
    INDEX
    Explanations

    if you or pregnant

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

    This neuron detects conditional “if you…” phrases used to give instructions or advice.

    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
     lada
    -1.46
     Boletín
    -1.34
    ïe
    -1.33
     Ukraina
    -1.31
    鲥
    -1.29
    tront
    -1.28
     charmant
    -1.27
     Wikip
    -1.27
     bensin
    -1.27
    wesen
    -1.25
    POSITIVE LOGITS
     still
    1.61
     not
    1.47
     more
    1.42
     have
    1.36
     noch
    1.35
     no
    1.34
     remove
    1.25
     excessive
    1.23
     over
    1.23
     nicht
    1.21
    Activations Density 0.020%

    No Known Activations