© 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. 3-GEMMASCOPE-2-TRANSCODER-262K
    7. 34595
    Prev
    Next
    INDEX
    Explanations

    hill climbing and locations

    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_3_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
    ,
    0.70
    ری
    0.66
     сообщи
    0.66
     позволи
    0.66
    ен
    0.64
    ພວກເຮ
    0.61
    ことにより
    0.61
     coalgebras
    0.61
     décerné
    0.60
    ピソード
    0.59
    POSITIVE LOGITS
     I
    0.83
     Hill
    0.69
    న
    0.62
    ya
    0.61
     hill
    0.61
    झा
    0.58
    inz
    0.57
     J
    0.56
     Jung
    0.55
     C
    0.54
    Activations Density 0.000%

    No Known Activations