© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27591
    Prev
    Next
    INDEX
    Explanations

    URLs and documentation

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

    The neuron fires on parts of web addresses—tokens like domain names, slashes, file‐path segments, and TLDs—indicating it’s looking for URL or URI patterns.

    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
     criatura
    -1.34
    AnchorStyles
    -1.33
    viève
    -1.31
     pierdas
    -1.29
    
    -1.25
    ParallelGroup
    -1.22
     indifer
    -1.22
     ľudí
    -1.21
     blessé
    -1.21
    幹線
    -1.20
    POSITIVE LOGITS
    Geschiedenis
    1.11
    躥
    1.10
    </em>
    1.10
    為に
    1.05
    yesterday
    1.05
    を選び
    1.05
     proposta
    1.04
     gratuitamente
    1.01
     gradualmente
    1.01
    مپ
    1.00
    Activations Density 0.034%

    No Known Activations