© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 15519
    Prev
    Next
    INDEX
    Explanations

    under certain conditions

    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
     تاکہ
    0.43
     ताकि
    0.42
    claration
    0.41
     safest
    0.38
    以免
    0.38
    还是要
    0.38
    เพื่อให้
    0.37
    举
    0.37
     чтобы
    0.36
     unenforceable
    0.36
    POSITIVE LOGITS
     properly
    1.46
     appropriately
    1.37
     suitably
    1.34
     Properly
    1.34
     sufficiently
    1.28
    適切な
    1.24
     certain
    1.22
    proper
    1.20
     megfelelő
    1.20
     правильно
    1.19
    Activations Density 0.046%

    No Known Activations