© 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. 130491
    Prev
    Next
    INDEX
    Explanations

    The neuron specializes in spotting numeric expressions (digits, decimal numbers, hyphens in numeric ranges, etc.).

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

    said or said that

    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
    icherung
    -1.00
    
    -0.99
     mori
    -0.94
    んでいる
    -0.93
     technics
    -0.92
     CLAIMS
    -0.91
     twit
    -0.91
     Smell
    -0.90
     cang
    -0.90
     HAPPENED
    -0.90
    POSITIVE LOGITS
     Орден
    1.06
     puesto
    0.96
    ̐
    0.95
     Amongst
    0.93
     distintas
    0.92
    Speaking
    0.89
     mówi
    0.89
     sabido
    0.89
    》（
    0.89
     Esta
    0.89
    Activations Density 0.009%

    No Known Activations