© 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. 6381
    Prev
    Next
    INDEX
    Explanations

    clarifying questions

    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
     guess
    0.35
    giers
    0.34
     повто
    0.33
     Hire
    0.33
     hire
    0.32
     tear
    0.32
    Albert
    0.32
    嘭
    0.32
     bowed
    0.32
     told
    0.32
    POSITIVE LOGITS
     hỏi
    0.48
     vragen
    0.48
     Question
    0.47
     question
    0.47
    Question
    0.47
     vraag
    0.47
     spør
    0.46
    问
    0.46
     clarification
    0.45
    question
    0.45
    Activations Density 0.009%

    No Known Activations