© 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. 107467
    Prev
    Next
    INDEX
    Explanations

    Q: questions phrased with Q:

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

    The neuron responds to long sequences of digits—that is, multi-digit 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
     their
    -1.03
     by
    -1.02
     its
    -0.97
     supports
    -0.90
     most
    -0.90
     well
    -0.90
     Unlike
    -0.87
     an
    -0.87
     at
    -0.86
     Only
    -0.86
    POSITIVE LOGITS
     "*"
    1.10
     Machu
    0.98
     MASK
    0.93
     Karol
    0.93
     mower
    0.90
    bew
    0.90
    Helm
    0.89
     acumen
    0.87
     iedereen
    0.86
     topaz
    0.85
    Activations Density 0.003%

    No Known Activations