© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 54310
    Prev
    Next
    INDEX
    Explanations

    after a certain word

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

    This neuron responds to personal pronouns—especially “you” (and to a lesser extent “I” and other first‐/second‐person references) indicating direct address.

    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
     bolster
    -1.68
     undeniably
    -1.55
     invigor
    -1.52
     exquisitely
    -1.52
     unrelenting
    -1.51
     five
    -1.48
     oddly
    -1.48
     four
    -1.48
     Surprisingly
    -1.48
     unusually
    -1.45
    POSITIVE LOGITS
     crucial
    1.73
     largely
    1.66
     reiterated
    1.61
     fabulous
    1.59
    氿
    1.59
     dando
    1.55
     alcuni
    1.53
     predominantly
    1.50
     dreadful
    1.49
     Dinas
    1.49
    Activations Density 0.038%

    No Known Activations