© 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. 23-GEMMASCOPE-2-TRANSCODER-262K
    7. 89086
    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_23_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
     раздра
    0.42
    PrintError
    0.40
    柴
    0.39
    ANCES
    0.39
    brit
    0.39
    Сан
    0.39
    bio
    0.38
     sofern
    0.38
    ської
    0.38
    ہائی
    0.38
    POSITIVE LOGITS
     Effective
    0.45
     effective
    0.41
     average
    0.39
     Abuja
    0.39
     Average
    0.37
     ayudar
    0.37
    ogiri
    0.36
    鹤
    0.36
     efectivos
    0.36
     Aberdeen
    0.36
    Activations Density 0.000%

    No Known Activations