© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129695
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens (digits or numbers) wherever they appear in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    levels or stages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     türlü
    -0.96
    wählen
    -0.85
    偻
    -0.84
    bottomrule
    -0.81
     kurios
    -0.80
    argmax
    -0.79
    Lastly
    -0.79
    verständ
    -0.79
    更是
    -0.79
    zelfde
    -0.79
    POSITIVE LOGITS
     next
    1.15
     intermediate
    1.00
     second
    0.92
     следующий
    0.80
     次
    0.79
     middle
    0.78
    intermediate
    0.77
     secondary
    0.75
     they
    0.75
     siguiente
    0.74
    Activations Density 0.077%

    No Known Activations