Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-16K
    4. 12543
    Prev
    Next
    INDEX
    Explanations

    However, instead, but

    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_31_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
     tentunya
    0.42
     虽然
    0.40
    もちろん
    0.39
    確かに
    0.39
     oczywiście
    0.38
     registros
    0.37
     relatório
    0.37
     registers
    0.37
     물론
    0.37
    レシピ
    0.37
    POSITIVE LOGITS
     BUT
    0.51
    有一种
    0.49
    했는데
    0.48
     그런데
    0.47
     ANOTHER
    0.46
    有时候
    0.46
    but
    0.45
     Occasionally
    0.43
     কিন্তু
    0.43
     occasionally
    0.43
    Activations Density 0.190%

    No Known Activations