© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 106967
    Prev
    Next
    INDEX
    Explanations

    Hacker News story posts

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

    The neuron detects parts of URLs and hyperlink markup (e.g., "https://", domain names, paths, query parameters, and surrounding brackets).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     EBITDA
    -0.77
     Doğ
    -0.74
     ды
    -0.74
    noteq
    -0.72
     Jefferson
    -0.72
     Durango
    -0.71
    zyden
    -0.71
     Jeans
    -0.71
    焜
    -0.70
     sider
    -0.68
    POSITIVE LOGITS
    story
    1.29
     stories
    1.21
     HN
    1.20
     story
    1.18
    Stories
    1.15
    Story
    1.14
    stories
    1.01
    HN
    1.01
     news
    1.00
    Hn
    1.00
    Activations Density 0.085%

    No Known Activations