© 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. 34-GEMMASCOPE-RES-131K
    7. 119344
    Prev
    Next
    INDEX
    Explanations

    works of media

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

    This neuron activates on numeric tokens—especially years and other multi-digit numbers—such as those found in filmography dates.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Wikipedia-style articles about actors and TV shows, particularly when listing filmography, TV series, or cast information.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    references to television and film productions, including series, shows, films, and role/credit listings in entertainment bios and titles.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     forbed
    -0.85
     Promoting
    -0.77
    Promoting
    -0.75
     שת
    -0.75
    grafica
    -0.73
    ıyorum
    -0.73
     Promote
    -0.73
    τσι
    -0.73
     Plastic
    -0.72
     Trailer
    -0.72
    POSITIVE LOGITS
     daunting
    0.77
     همس
    0.76
     commonly
    0.74
     arguably
    0.73
    adaan
    0.73
    Sesame
    0.73
     sogni
    0.71
     likely
    0.71
     mandal
    0.71
     Lodging
    0.71
    Activations Density 0.038%

    No Known Activations