© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6698
    Prev
    Next
    INDEX
    Explanations

    color and scrollbar styles

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

    This neuron fires specifically on standalone decimal‐formatted numbers (i.e. tokens composed of digits with a dot).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    0.87
     (“
    0.84
     Curves
    0.84
     Continuing
    0.83
     Centuries
    0.82
    
    0.82
    Bankruptcy
    0.80
     Communism
    0.80
     ভাবে
    0.80
     Chairs
    0.78
    POSITIVE LOGITS
    ol
    0.89
    ojen
    0.85
    ar
    0.82
    카
    0.80
    ר
    0.78
    你们
    0.77
     Jako
    0.76
     događ
    0.75
    q
    0.75
    iyar
    0.74
    Activations Density 0.000%

    No Known Activations