© 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. 10-GEMMASCOPE-2-TRANSCODER-262K
    7. 25631
    Prev
    Next
    INDEX
    Explanations

    sub- words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_10_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
     hydraulic
    0.71
     rotary
    0.68
     personas
    0.63
     migration
    0.62
     resolution
    0.61
     kitty
    0.61
    {
    0.61
     whirlpool
    0.60
     mechanism
    0.59
    جار
    0.59
    POSITIVE LOGITS
    o
    1.09
    ta
    0.96
    il
    0.91
    ad
    0.88
    ado
    0.87
    ce
    0.86
    ne
    0.84
    is
    0.84
    al
    0.83
    ING
    0.82
    Activations Density 0.010%

    No Known Activations