© 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. 252760
    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
    s
    0.45
    ochet
    0.39
     chromedp
    0.39
     fub
    0.37
    !!");
    0.37
    InitFlag
    0.36
    YL
    0.36
     invitado
    0.36
    ाइज
    0.35
    CLN
    0.35
    POSITIVE LOGITS
    рг
    0.40
     coaches
    0.39
     V
    0.38
    atures
    0.38
    asına
    0.38
    ceptive
    0.37
    ToOne
    0.36
    ุต
    0.35
    ()\
    0.35
     coach
    0.34
    Activations Density 0.000%

    No Known Activations