© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-262K
    4. 24141
    Prev
    Next
    INDEX
    Explanations

    usermediatypesmovieTVgamecontentmusicaudiovideoshortshort filmmovie trailerlive broadcastvideo gameshort film directorshort film festival</TOP_POSITIVE_LOGITS><MAX_ACTIVATING_TOKENS>videofilmsTV</MAX_ACTIVATING_TOKENS><TOP_ACTIVATING_TEXTS>movie trailer,media typeMedia typeIn the US, a "short film" is a film of 40 minutes or less, including credits.The documentary films won 4 awards.As a full-time student, you can get access and use of a wide range of media.With this technology you can add rich media to your web pages.The video games are a lot of fun.You can get access to various media.You can start your own TV channel.Many people are interested in movies, music, TV, and games.A TV show.In general, a \"short film\" is a film of 40 minutes or less, including credits.The film festival is a place to watch short films.You will be able to get access to a lot of media, the user can control the playback of the video.The new movie was released yesterday.</TOP_ACTIVATING_TEXTS><TOKENS_AFTER_MAX_ACTIVATING_TOKEN>gametypeproductiongamesgamesgamesgamesgamesgames</TOKENS_AFTER_MAX_ACTIVATING_TOKEN>Explain the neuron above with a word or phrase, not a complete sentence.media types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ఎస్‌
    0.49
    НЫ
    0.45
    ântul
    0.45
    krét
    0.45
    VARIABLES
    0.44
    рем
    0.43
     स्टार्टिंग
    0.42
     deteriorating
    0.42
    鏖
    0.42
     Variablen
    0.41
    POSITIVE LOGITS
     Michelangelo
    0.39
     electron
    0.37
     inspection
    0.36
    Mediterranean
    0.36
     fool
    0.36
    glue
    0.35
     inspections
    0.34
     fi
    0.34
     fí
    0.34
    features
    0.34
    Activations Density 0.001%

    No Known Activations