© 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. 46-GEMMASCOPE-2-TRANSCODER-262K
    7. 203514
    Prev
    Next
    INDEX
    Explanations

    gold or clean

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_46_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
     bad
    1.11
    bad
    0.88
     Bad
    0.86
     BAD
    0.85
    Bad
    0.84
    BAD
    0.75
     بد
    0.69
    坏
    0.65
     bada
    0.62
    壞
    0.54
    POSITIVE LOGITS
     Gold
    0.43
    Gold
    0.43
    Clean
    0.40
    まれ
    0.39
     गोल्ड
    0.38
    Mil
    0.37
     clean
    0.37
     प्रदूषण
    0.37
     הצ
    0.37
     Clean
    0.36
    Activations Density 0.093%

    No Known Activations