© 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-16K
    7. 9120
    Prev
    Next
    INDEX
    Explanations

    asking about actions

    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_16k_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
    IRY
    0.55
     from
    0.50
     terms
    0.46
     Issues
    0.46
     comes
    0.46
    ==
    0.45
     kommer
    0.45
     مسائل
    0.45
    resa
    0.44
     Manners
    0.44
    POSITIVE LOGITS
     действий
    0.53
    ларга
    0.48
    Twig
    0.47
     કાર્યવાહી
    0.45
    Done
    0.44
     noviembre
    0.44
    ِين
    0.44
    <0xA5>
    0.42
     azioni
    0.42
     सराह
    0.42
    Activations Density 0.200%

    No Known Activations