© 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. 30-GEMMASCOPE-2-TRANSCODER-262K
    7. 53143
    Prev
    Next
    INDEX
    Explanations

    explain why harmful requests are refused

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_30_width_262k_l0_small_affine
    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
     détach
    0.27
    这么多
    0.25
    肄
    0.25
    ্বর
    0.24
     ovviamente
    0.24
     Composite
    0.24
    Composite
    0.24
    章
    0.24
    composite
    0.24
    不仅仅
    0.24
    POSITIVE LOGITS
     leis
    0.30
    abola
    0.27
     pregnancies
    0.27
     migraines
    0.26
     Megan
    0.26
     आपके
    0.25
    𝘼
    0.25
     regres
    0.25
     bags
    0.25
     Priya
    0.25
    Activations Density 0.002%

    No Known Activations