© 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. 39-GEMMASCOPE-2-TRANSCODER-262K
    7. 5243
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    ǁ
    0.45
     دریا
    0.39
    憸
    0.38
    আত
    0.36
     ಪ್ರಮಾಣ
    0.36
    циона
    0.35
    កំណ
    0.35
    тинен
    0.35
     coherent
    0.35
    阡
    0.35
    POSITIVE LOGITS
    Buffalo
    0.37
    pr
    0.37
    eng
    0.37
     Buffalo
    0.37
    Duel
    0.34
    wood
    0.34
    ቻል
    0.34
    溶
    0.34
    sp
    0.34
     Medizin
    0.34
    Activations Density 0.000%

    No Known Activations