© 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. 133662
    Prev
    Next
    INDEX
    Explanations

    km

    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
     N
    0.45
     nain
    0.42
     Orc
    0.40
    N
    0.40
     الن
    0.39
    어나
    0.39
    HARAD
    0.39
     Calais
    0.38
    FlatAppearance
    0.38
    redos
    0.37
    POSITIVE LOGITS
     TU
    0.61
    T
    0.60
     t
    0.59
     тъ
    0.59
     tumb
    0.57
     T
    0.57
     tamb
    0.56
     tam
    0.56
     Tamb
    0.56
     ту
    0.55
    Activations Density 0.000%

    No Known Activations