© 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. 56-GEMMASCOPE-2-TRANSCODER-262K
    7. 205102
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_56_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
     comparing
    0.42
    umuza
    0.41
     exploring
    0.40
     renting
    0.37
    啾
    0.37
     resisting
    0.37
    orne
    0.37
    bersihan
    0.37
     falling
    0.36
     flying
    0.36
    POSITIVE LOGITS
     vraiment
    0.43
    really
    0.43
    Really
    0.40
     όπως
    0.39
    브
    0.39
    เช่น
    0.39
    샘
    0.38
     really
    0.38
     davvero
    0.38
     satisf
    0.37
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.