© 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. 40-GEMMASCOPE-2-RES-65K
    7. 2015
    Prev
    Next
    INDEX
    Explanations

    try except

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_65k_l0_medium
    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
    เมริกา
    0.42
    =\""
    0.42
    渠
    0.42
    ึ่ง
    0.42
    阈
    0.41
     সংজ্ঞা
    0.40
    ivez
    0.39
    函數
    0.38
    衄
    0.38
    infodisc
    0.38
    POSITIVE LOGITS
     catch
    1.60
     catches
    1.47
     catching
    1.46
     Catch
    1.42
    catch
    1.39
    Catch
    1.34
    catching
    1.13
     caught
    1.07
     try
    1.02
    try
    0.98
    Activations Density 0.331%

    No Known Activations