© 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. 10-GEMMASCOPE-RES-131K
    7. 85174
    Prev
    Next
    INDEX
    Explanations

    cdn URLs

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

    The neuron is primarily firing on URL‐related “cdn” tokens (and similar low-frequency identifier strings), i.e. bits of content that look like content‐delivery network or domain-name markers rather than ordinary prose.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    »),
    -2.61
    isation
    -2.39
    e
    -2.36
    鯪
    -2.27
    These
    -2.27
    。“
    -2.23
    ificado
    -2.22
    Ⲉ
    -2.20
    umplimiento
    -2.19
    …»
    -2.17
    POSITIVE LOGITS
    錞
    2.48
     Of
    2.39
     wichtigste
    2.34
    跸
    2.34
     wonderfully
    2.33
     —
    2.33
    ௡
    2.28
    '
    2.27
     citing
    2.20
     gazing
    2.20
    Activations Density 0.001%

    No Known Activations