© 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. 15589
    Prev
    Next
    INDEX
    Explanations

    natural phrasing, certain connotations, disclaimer

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

    text-structuring and formatting cues—like headings, options, and list-style organization often used in instructional or markdown-formatted responses.

    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_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
    דות
    0.46
    ኹ
    0.46
    ﾈ
    0.45
    也就
    0.45
     nöt
    0.44
     ווא
    0.42
    carbam
    0.41
    埤
    0.41
    chant
    0.40
    ्‍य
    0.40
    POSITIVE LOGITS
    0.46
     पैक
    0.45
     Drawer
    0.43
    with
    0.43
    లతో
    0.43
     Network
    0.41
     với
    0.40
     Kits
    0.40
     Lorsque
    0.39
    iril
    0.39
    Activations Density 23.369%

    No Known Activations