© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 56-GEMMASCOPE-2-TRANSCODER-262K
    7. 251823
    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
     Disabilities
    0.39
     Gales
    0.38
     SWAT
    0.36
     th
    0.36
     azt
    0.36
     myriad
    0.36
     eram
    0.36
     Г
    0.36
     SV
    0.36
    ادی
    0.35
    POSITIVE LOGITS
    R
    0.43
    Passed
    0.41
    workflow
    0.40
    effect
    0.39
    Effect
    0.39
    passed
    0.38
    𝑅
    0.38
    RR
    0.38
    Yon
    0.38
    mental
    0.38
    Activations Density 0.000%

    No Known Activations