© 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. 39-GEMMASCOPE-2-TRANSCODER-262K
    7. 1247
    Prev
    Next
    INDEX
    Explanations

    I'm, I was, I wanted

    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_39_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
    being
    0.33
     سیم
    0.33
    enangan
    0.33
    ారం
    0.33
    once
    0.32
     tuyển
    0.32
    忿
    0.32
     конкур
    0.32
    hebung
    0.32
     lacks
    0.32
    POSITIVE LOGITS
     gave
    0.35
    ŕ
    0.35
     sen
    0.35
    чие
    0.35
    elli
    0.35
     Sen
    0.33
     sustained
    0.33
    gee
    0.32
     কাহারও
    0.32
    Rw
    0.32
    Activations Density 0.141%

    No Known Activations