© 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. 40-GEMMASCOPE-2-RES-16K
    7. 11734
    Prev
    Next
    INDEX
    Explanations

    trails and dirt courses

    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_16k_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
     الطريق
    0.38
     cyclists
    0.38
    リエステル
    0.38
     नेहरू
    0.37
    樯
    0.36
     feedwater
    0.36
    াদিগের
    0.36
    恒
    0.35
     সিরাজ
    0.35
     Meteorology
    0.35
    POSITIVE LOGITS
     trail
    0.67
     Trail
    0.64
     Dirt
    0.63
     dirt
    0.59
     Trials
    0.58
     enduro
    0.58
     trails
    0.57
     Trails
    0.57
    Trial
    0.57
    Dirt
    0.57
    Activations Density 0.043%

    No Known Activations