© 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-3-27B-IT
    5. 50-GEMMASCOPE-2-TRANSCODER-262K
    7. 213401
    Prev
    Next
    INDEX
    Explanations

    music facts

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_50_width_262k_l0_small_affine
    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
     mashed
    0.38
     mus
    0.38
    क्तिक
    0.38
     Soma
    0.37
     Bass
    0.36
    |^{\
    0.36
     バス
    0.36
    বিন
    0.36
     perif
    0.35
     спи
    0.35
    POSITIVE LOGITS
    荃
    0.43
    resident
    0.41
    pport
    0.40
    ajo
    0.40
    লাইয়া
    0.40
     Julio
    0.39
    emiah
    0.39
    ätten
    0.38
     Fact
    0.38
     FACT
    0.37
    Activations Density 0.003%

    No Known Activations