© 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. 10-GEMMASCOPE-RES-131K
    7. 71868
    Prev
    Next
    INDEX
    Explanations

    "this" followed by content item

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

    The neuron specializes in spotting call-to-action engagement prompts (e.g. “If you like this story…,” “found it useful,” “share this”) that invite readers to interact with or share the content.

    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
    ”？
    -1.41
    ඈ
    -1.35
     remnants
    -1.20
    LECTIONS
    -1.14
    嫉妒
    -1.13
    よろしくお願い
    -1.11
     universidades
    -1.09
     often
    -1.09
     dört
    -1.09
     problemática
    -1.08
    POSITIVE LOGITS
    either
    1.22
    łby
    1.19
    рили
    1.18
    1
    1.16
     Berechnung
    1.15
    讓
    1.13
    kasa
    1.13
    这个
    1.12
    licula
    1.10
     then
    1.09
    Activations Density 0.033%

    No Known Activations