© 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. 2-GEMMASCOPE-2-TRANSCODER-262K
    7. 61158
    Prev
    Next
    INDEX
    Explanations

    pillars

    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_2_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
    dır
    1.42
    は
    1.39
    ম
    1.34
    는
    1.29
    리
    1.18
    ार
    1.16
    ла
    1.14
    ა
    1.14
    ли
    1.12
    ك
    1.09
    POSITIVE LOGITS
     to
    0.94
     increases
    0.90
     encrypted
    0.88
    ants
    0.87
     affective
    0.86
     ingredient
    0.84
    </h2>
    0.83
     executed
    0.82
    <0x0D>
    0.82
     ingredients
    0.82
    Activations Density 0.000%

    No Known Activations