© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 2802
    Prev
    Next
    INDEX
    Explanations

    exploring curiosity or interest

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

    sentences offering help or crisis-resource language addressed to the user (e.g., "If you are..." / "you are...") in safety/refusal responses.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     মোটামুটি
    0.49
    简直
    0.43
     डिग्री
    0.40
     waiters
    0.39
     peanuts
    0.39
     בעי
    0.38
    !">
    0.37
     delicios
    0.37
     specialize
    0.37
     downright
    0.37
    POSITIVE LOGITS
     curiosity
    0.80
    寻求
    0.70
     관심
    0.63
     interesse
    0.63
     जिज्ञासा
    0.62
     관심을
    0.62
    好奇
    0.61
     seeking
    0.61
     seek
    0.60
     interest
    0.59
    Activations Density 0.068%

    No Known Activations