© 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. 153364
    Prev
    Next
    INDEX
    Explanations

    stretching, stretched

    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_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
     washes
    0.46
     climbs
    0.41
     superseded
    0.40
    리아
    0.39
     delineation
    0.39
    Ǎ
    0.39
    ^_
    0.39
     Yon
    0.39
    Ձ
    0.38
     synthesis
    0.38
    POSITIVE LOGITS
     Reddit
    0.43
     หรือ
    0.41
     Net
    0.41
    reddit
    0.41
    Reddit
    0.40
     bár
    0.39
    bbe
    0.38
     reddit
    0.38
    Net
    0.38
     net
    0.36
    Activations Density 0.004%

    No Known Activations