© 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. 11337
    Prev
    Next
    INDEX
    Explanations

    Retail Clinics, yogurt is, Code Generation

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

    mentions of proper nouns and title-cased terms—like brand, product, or technical names—especially when used in headings or emphasized phrases.

    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
     irritating
    0.42
    的样子
    0.38
     hon
    0.37
    ન
    0.37
     unpopular
    0.36
     sores
    0.35
     মোটামুটি
    0.35
    હ
    0.35
     originals
    0.35
     அக
    0.34
    POSITIVE LOGITS
    による
    0.50
    也能
    0.47
    を活用
    0.41
     offrant
    0.40
     utilisant
    0.40
    裡的
    0.39
    Instead
    0.38
     brazo
    0.38
    作為
    0.38
    으로서
    0.38
    Activations Density 0.347%

    No Known Activations