© 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. 25-GEMMASCOPE-2-TRANSCODER-262K
    7. 237911
    Prev
    Next
    INDEX
    Explanations

    interpreting user needs

    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_25_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
    نے
    0.43
    ২
    0.42
     három
    0.41
    تى
    0.41
    мут
    0.40
     clamping
    0.39
    কেন্ড
    0.38
     پھر
    0.38
     تط
    0.38
    પર
    0.38
    POSITIVE LOGITS
     Cre
    0.43
     Mark
    0.39
     Kh
    0.39
    imenta
    0.38
     M
    0.37
     Stephen
    0.37
     White
    0.37
     corpus
    0.37
     sent
    0.36
     constitution
    0.36
    Activations Density 0.006%

    No Known Activations