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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-262K
    4. 138485
    Prev
    Next
    INDEX
    Explanations

    Leo Maxwell hugging people

    np_acts-logits-general · gemini-2.5-flash-lite

    narrative and outline structure elements in responses, especially section headers, numbered list items, and prominent character name mentions.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
     renta
    0.53
    ੀ
    0.52
     constitución
    0.52
     contratación
    0.52
     ಪ್ರಾರಂಭ
    0.50
     través
    0.50
     рис
    0.48
     propaganda
    0.47
     critica
    0.47
     transacción
    0.47
    POSITIVE LOGITS
     Break
    0.47
    ,
    0.45
     Workspace
    0.44
     Amended
    0.44
     Wrestling
    0.43
     Summers
    0.42
     Relations
    0.42
     Affairs
    0.41
     Quarter
    0.41
    お手
    0.40
    Activations Density 0.008%

    No Known Activations