© 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-262K
    7. 194844
    Prev
    Next
    INDEX
    Explanations

    writing notes about importance

    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_262k_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
    ?:
    0.41
    ▔
    0.41
     ধর্ষ
    0.41
    勹
    0.40
     parlé
    0.39
     dépenses
    0.39
    ┕
    0.38
    フェ
    0.37
     seul
    0.37
    fei
    0.37
    POSITIVE LOGITS
     lordship
    0.41
    dumbbell
    0.40
     देकर
    0.40
     Hs
    0.38
     Pilgrim
    0.37
     pilgrims
    0.36
     Lloyds
    0.36
     عدالت
    0.36
     സർക്കാർ
    0.36
     Bey
    0.35
    Activations Density 0.110%

    No Known Activations