© 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. 22673
    Prev
    Next
    INDEX
    Explanations

    greetings followed by comma

    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
    いる
    0.54
    지를
    0.47
    在
    0.46
    도
    0.45
    的方式
    0.45
    irmer
    0.45
    ாது
    0.44
    丸
    0.44
     twig
    0.44
    生命
    0.43
    POSITIVE LOGITS
     Muslims
    0.53
     uproar
    0.51
     Operation
    0.44
    vann
    0.43
    izacji
    0.43
     .”
    0.42
    Crazy
    0.42
    finaly
    0.42
     Afghans
    0.42
     Afrika
    0.42
    Activations Density 0.002%

    No Known Activations