© 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. 41-GEMMASCOPE-2-RES-262K
    7. 64356
    Prev
    Next
    INDEX
    Explanations

    added or increased things

    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_all/layer_41_width_262k_l0_big
    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
     impatient
    0.43
     быстрее
    0.43
     দ্রুত
    0.41
    缺少
    0.41
    Scala
    0.39
    髙
    0.39
     деше
    0.38
    경
    0.38
    굶
    0.38
     скоро
    0.37
    POSITIVE LOGITS
     thêm
    0.97
    增加了
    0.97
    额外的
    0.94
     increased
    0.91
    增加
    0.89
     menambah
    0.88
     extra
    0.87
     adds
    0.86
     zusätzlichen
    0.86
    额外
    0.84
    Activations Density 0.000%

    No Known Activations