© 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. 22-GEMMASCOPE-RES-131K
    7. 69092
    Prev
    Next
    INDEX
    Explanations

    years and versions

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

    This neuron activates on tokens containing numeric characters (digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    这句话
    -1.08
    
    -1.00
    quirer
    -1.00
    
    -0.95
    槼
    -0.94
    臜
    -0.94
    
    -0.90
    菒
    -0.90
    昨年
    -0.90
     gav
    -0.89
    POSITIVE LOGITS
     their
    0.89
     at
    0.86
     ubica
    0.84
     eats
    0.82
     Росси
    0.77
    пада
    0.77
     through
    0.76
    Researcher
    0.76
    hljs
    0.75
    huang
    0.75
    Activations Density 0.006%

    No Known Activations