© 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. 34-GEMMASCOPE-RES-131K
    4. 59934
    Prev
    Next
    INDEX
    Explanations

    obtaining permissions

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

    The neuron is highly responsive to non-narrative “metadata” tokens—things like embedded file or section labels, numeric IDs, headings, and other boilerplate markers rather than ordinary prose words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     around
    -0.88
     either
    -0.88
     obyvateľov
    -0.85
     Was
    -0.85
     umsch
    -0.84
    Feb
    -0.79
     had
    -0.79
     perhaps
    -0.77
     about
    -0.76
     while
    -0.76
    POSITIVE LOGITS
     thesis
    1.20
     willy
    1.06
     amis
    0.96
    Ayr
    0.93
     agres
    0.91
     Thesis
    0.91
    prache
    0.90
     sows
    0.90
     theses
    0.90
     Kills
    0.90
    Activations Density 0.000%

    No Known Activations