© 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. 31-GEMMASCOPE-2-RES-262K
    7. 7602
    Prev
    Next
    INDEX
    Explanations

    incentives for improvement

    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
     Temperatura
    0.35
    毁灭
    0.35
    误差
    0.34
     herhangi
    0.32
    陨
    0.32
     trứng
    0.32
    மற்ற
    0.32
    亣
    0.31
    极限
    0.31
    单词
    0.31
    POSITIVE LOGITS
     incentives
    0.56
    incent
    0.54
     incentiv
    0.52
     education
    0.52
     incentivize
    0.50
     improved
    0.50
     incentive
    0.50
     Incent
    0.48
     improve
    0.47
     partnerships
    0.47
    Activations Density 0.129%

    No Known Activations