© 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. 10-GEMMASCOPE-RES-131K
    7. 120162
    Prev
    Next
    INDEX
    Explanations

    start of movie and tv show titles

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

    The neuron strongly activates on the opening quotation mark (and immediately following token) of a titled work—i.e. the beginning of a film, play, book, or article title.

    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
     
    -2.20
     new
    -1.95
     has
    -1.79
     out
    -1.78
     in
    -1.70
     $
    -1.67
     highly
    -1.66
     sizable
    -1.60
     intricate
    -1.60
     .
    -1.59
    POSITIVE LOGITS
     schuldig
    1.98
    ("")
    1.74
     电视
    1.74
     créer
    1.69
     senza
    1.69
     alebo
    1.66
    äld
    1.66
    😾
    1.66
     viſ
    1.63
     veya
    1.63
    Activations Density 0.015%

    No Known Activations