© 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. 31-GEMMASCOPE-2-TRANSCODER-262K
    7. 50460
    Prev
    Next
    INDEX
    Explanations

    step away, step up, step back

    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_31_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
    用心
    0.37
    読
    0.37
     exponential
    0.37
     SeekBar
    0.37
     Verbal
    0.37
     tiến
    0.36
     inherit
    0.35
     उच्चारण
    0.35
    鉚
    0.34
     Matthias
    0.34
    POSITIVE LOGITS
     avoids
    0.36
     replaces
    0.35
     badi
    0.35
    Ge
    0.34
     involves
    0.34
     complies
    0.34
     partito
    0.34
     covers
    0.33
     compares
    0.33
     apart
    0.33
    Activations Density 0.002%

    No Known Activations