© 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-16K
    7. 10035
    Prev
    Next
    INDEX
    Explanations

    way to or way of

    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
    amaz
    0.44
    وئ
    0.41
    OrEqualTo
    0.41
    rockets
    0.40
     Amaz
    0.40
     решению
    0.39
    ब्राह
    0.39
     ។
    0.39
    Rocket
    0.39
    Amaz
    0.39
    POSITIVE LOGITS
     way
    0.49
     subconsciously
    0.41
     save
    0.38
     dumb
    0.38
     måde
    0.38
     visually
    0.38
     reclaim
    0.37
     easily
    0.37
     bridge
    0.37
     regain
    0.37
    Activations Density 0.043%

    No Known Activations