© 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. 76
    Prev
    Next
    INDEX
    Explanations

    website domain names

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

    The neuron detects fragments of web addresses and domain names (e.g. “.com,” “.net,” “EpisodeDate,” URL-like tokens).

    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
     at
    -1.75
     both
    -1.73
     then
    -1.63
     their
    -1.61
     first
    -1.57
     on
    -1.49
     six
    -1.48
     eight
    -1.45
     three
    -1.41
     seven
    -1.41
    POSITIVE LOGITS
     grotte
    1.45
     Dinas
    1.45
    trying
    1.42
     faciliter
    1.42
     assister
    1.38
    could
    1.37
    킁
    1.37
     COULD
    1.34
    almost
    1.32
    somewhat
    1.32
    Activations Density 0.006%

    No Known Activations