© 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. 55060
    Prev
    Next
    INDEX
    Explanations

    start of response or transition

    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
    শোধ
    0.53
     decay
    0.46
    viendo
    0.44
     demás
    0.44
    જનો
    0.43
    ysuckle
    0.42
    ᱢ
    0.42
     vykon
    0.42
     pouvoirs
    0.41
    ▔
    0.41
    POSITIVE LOGITS
     человеку
    0.47
     нашим
    0.46
    ↵↵
    0.44
    คุณ
    0.43
     centrally
    0.43
     ограничи
    0.43
    ó
    0.43
     you
    0.43
     "
    0.43
     car
    0.42
    Activations Density 0.003%

    No Known Activations