© 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. 17747
    Prev
    Next
    INDEX
    Explanations

    when dealing with

    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
     यू
    0.38
     maka
    0.37
     ហើយ
    0.37
    intaan
    0.37
     संसार
    0.36
     tí
    0.35
     этому
    0.34
    䒠
    0.34
     கூறி
    0.34
     tejto
    0.34
    POSITIVE LOGITS
     dealing
    1.08
    dealing
    0.92
     Dealing
    0.89
    Dealing
    0.83
     التعامل
    0.61
    扱う
    0.59
     речь
    0.57
     deal
    0.56
     dealt
    0.56
     Deal
    0.52
    Activations Density 0.010%

    No Known Activations