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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-262K
    4. 36629
    Prev
    Next
    INDEX
    Explanations

    contractual provisions, system tools, thread IDs, dataframes

    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_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
     મળશે
    0.54
     принадле
    0.52
     помогут
    0.49
     соотно
    0.47
     фина
    0.47
     серви
    0.47
     იქნება
    0.46
     podrás
    0.46
     फिलहाल
    0.45
     получите
    0.45
    POSITIVE LOGITS
     was
    0.55
     était
    0.52
     quelquefois
    0.51
     damals
    0.46
    当时
    0.46
    이었다
    0.45
     xvi
    0.45
     had
    0.45
    Elegant
    0.44
     столь
    0.44
    Activations Density 0.009%

    No Known Activations