© 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. 204825
    Prev
    Next
    INDEX
    Explanations

    risks and negative outcomes

    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
     stamped
    0.50
    throws
    0.47
     hairstyle
    0.46
     spiritually
    0.45
     stamping
    0.44
    लै
    0.44
    蝶
    0.44
     hairstyles
    0.43
     demonic
    0.43
     dalle
    0.43
    POSITIVE LOGITS
    Relative
    0.45
     과학
    0.44
     ਲ
    0.44
     обу
    0.43
    ONES
    0.43
    Ethiop
    0.43
     โอ
    0.42
     зазна
    0.42
    oleh
    0.42
    Arti
    0.42
    Activations Density 0.000%

    No Known Activations