© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 20904
    Prev
    Next
    INDEX
    Explanations

    refusal response from model

    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
     foreshadow
    0.43
     upheaval
    0.40
     Ile
    0.39
    ynes
    0.38
     uphe
    0.38
    influenced
    0.38
     succ
    0.37
    <unused2007>
    0.37
     influenced
    0.37
     ஏற்படுத்த
    0.36
    POSITIVE LOGITS
    は
    0.51
    mming
    0.47
    mpeg
    0.46
    는
    0.46
     মূলত
    0.46
    本日
    0.45
     맛있
    0.43
    生素
    0.43
    是以
    0.43
    們
    0.43
    Activations Density 0.008%

    No Known Activations