© 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. 46474
    Prev
    Next
    INDEX
    Explanations

    determine sets, heavily influenced

    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
    is
    0.68
    ر
    0.60
    다
    0.59
    ب
    0.57
     każdy
    0.52
    k
    0.52
    ാത്ത
    0.51
    সাধারণ
    0.50
    கொ
    0.50
    Salem
    0.50
    POSITIVE LOGITS
     yc
    0.50
     gemäß
    0.49
    <0x92>
    0.47
     för
    0.47
     hypothetical
    0.46
     Appliances
    0.45
    <0x8A>
    0.45
    redditmedia
    0.45
     nutty
    0.44
     asparagus
    0.44
    Activations Density 0.000%

    No Known Activations