© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 9597
    Prev
    Next
    INDEX
    Explanations

    URLs and web addresses

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

    The neuron detects the hostnames of web URLs (the domain names, e.g. “google”, “microsoft”, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    を
    -2.22
    は
    -2.17
    が
    -2.11
    などの
    -1.99
     
    -1.98
     in
    -1.97
    '
    -1.91
     "
    -1.90
    ti
    -1.89
    も
    -1.85
    POSITIVE LOGITS
     and
    2.55
     delicada
    2.28
    曩
    2.20
    敍
    2.13
    んですよね
    2.13
    Зно
    2.09
    をします
    2.09
    のお話
    2.05
    ようで
    2.03
    誰が
    2.03
    Activations Density 0.066%

    No Known Activations