© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 166052
    Prev
    Next
    INDEX
    Explanations

    tasks. High quality

    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_40_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
     twelve
    0.41
     १२
    0.40
     Niagara
    0.40
     Waterford
    0.39
    wells
    0.39
     terminé
    0.39
    |)
    0.38
     tof
    0.38
    weiten
    0.37
     möglich
    0.37
    POSITIVE LOGITS
    edict
    0.39
    智慧
    0.38
     plainly
    0.37
    Ӯ
    0.36
    عنوان
    0.36
     wise
    0.35
     सीनियर
    0.35
     έχει
    0.35
    வதன்
    0.35
    ައި
    0.35
    Activations Density 0.001%

    No Known Activations