© 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. 205350
    Prev
    Next
    INDEX
    Explanations

    roles, characteristics, or components

    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
    ی
    0.53
     Taiwan
    0.49
    Ori
    0.47
    Assist
    0.45
    "]
    0.45
     व्ही
    0.43
     हवाई
    0.43
     പരിശീല
    0.43
     दवाई
    0.42
     महाराष्ट्र
    0.42
    POSITIVE LOGITS
     consecrated
    0.59
     exige
    0.48
     объема
    0.48
    листы
    0.48
    }']
    0.47
     списке
    0.46
     engulfed
    0.45
     labours
    0.45
     exotic
    0.45
     coloured
    0.45
    Activations Density 0.001%

    No Known Activations