© 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. 155032
    Prev
    Next
    INDEX
    Explanations

    messages

    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
     Ме
    0.52
     メ
    0.43
     মে
    0.43
     ме
    0.43
     μέ
    0.42
     ಮೆ
    0.42
    Ме
    0.41
     масса
    0.41
     masse
    0.40
     메
    0.40
    POSITIVE LOGITS
     messages
    1.16
    messages
    1.09
     Nachrichten
    1.05
     Messages
    1.02
     сообщения
    1.02
    Messages
    0.99
     Nachricht
    0.96
    消息
    0.95
     mensajes
    0.93
     messaging
    0.92
    Activations Density 0.004%

    No Known Activations