© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 19795
    Prev
    Next
    INDEX
    Explanations

    still continuing despite

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

    sentences or phrases that state safety rules, disclaimers, refusals to follow harmful instructions, or otherwise discuss model policies/ethical limits.

    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_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
    arant
    0.40
     setCurrent
    0.38
     actuellement
    0.36
    かと
    0.36
     혹은
    0.36
     현재
    0.35
    乆
    0.35
    음에
    0.34
     internationalen
    0.34
    іга
    0.34
    POSITIVE LOGITS
     still
    1.64
    それでも
    1.50
    still
    1.49
     STILL
    1.44
    仍然
    1.42
    Still
    1.41
     vẫn
    1.41
     Still
    1.41
     trotzdem
    1.39
     nonetheless
    1.36
    Activations Density 0.239%

    No Known Activations