© 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. 34-GEMMASCOPE-RES-131K
    7. 38637
    Prev
    Next
    INDEX
    Explanations

    .orig file backups

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

    The neuron fires on numeric tokens (including version numbers, dates, line‐counts, and other purely numeric strings).

    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
     closer
    -0.77
     والق
    -0.74
    ksamkeit
    -0.72
    ґ
    -0.70
     kirj
    -0.69
    本の
    -0.69
    веря
    -0.69
     फिर
    -0.67
     conjunction
    -0.66
    URIS
    -0.66
    POSITIVE LOGITS
    bak
    1.09
    BAK
    1.05
     backup
    0.92
     méta
    0.90
    .%
    0.89
     bak
    0.88
     tmp
    0.88
    eit
    0.87
     BAK
    0.87
    Bak
    0.86
    Activations Density 0.107%

    No Known Activations