© 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. 7589
    Prev
    Next
    INDEX
    Explanations

    requesting to produce a

    np_acts-logits-general · gemini-2.5-flash-lite

    explicit user instructions or help-seeking prompts requesting the assistant to perform a task, provide information, or give guidance.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
     λόγω
    0.27
    kannten
    0.24
     were
    0.23
     hạn
    0.23
    °.
    0.23
    infrastructure
    0.23
     কারণ
    0.23
    exists
    0.23
     অনেকে
    0.23
     addirittura
    0.23
    POSITIVE LOGITS
     пожалуйста
    0.29
    旅游
    0.28
    ﭑ
    0.27
     hairstyle
    0.27
     코
    0.26
     فراہم
    0.26
     написать
    0.26
     میکنم
    0.26
     witty
    0.26
     يساعد
    0.26
    Activations Density 1.695%

    No Known Activations