© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 127387
    Prev
    Next
    INDEX
    Explanations

    URL codes or identifiers

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

    The neuron fires on the short alphanumeric identifier tokens (e.g. pastebin or file‐sharing IDs) that appear as URL path or query fragments.

    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
     on
    -2.72
     at
    -2.59
     during
    -2.39
     of
    -2.34
     in
    -2.25
     to
    -2.25
     as
    -2.22
     should
    -2.19
     for
    -2.16
     before
    -2.14
    POSITIVE LOGITS
     ſever
    2.03
    謦
    1.95
    those
    1.91
    two
    1.90
    StoryboardSegue
    1.90
    﨑
    1.84
    agerie
    1.84
    ಱ
    1.84
    three
    1.83
    six
    1.82
    Activations Density 0.054%

    No Known Activations