© 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-65K
    7. 51741
    Prev
    Next
    INDEX
    Explanations

    easily distracted or fragile

    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
    峹
    0.41
    órmula
    0.37
    мини
    0.37
    ার্চ
    0.37
    烀
    0.36
     губер
    0.36
     neglected
    0.34
    éor
    0.34
    恜
    0.34
    Billing
    0.33
    POSITIVE LOGITS
     easily
    1.60
     Easily
    1.48
    easily
    1.45
     легко
    1.37
     facilmente
    1.37
     facilement
    1.30
     fragile
    1.27
     आसानी
    1.27
     fácilmente
    1.23
     সহজেই
    1.20
    Activations Density 0.137%

    No Known Activations