© 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. 4433
    Prev
    Next
    INDEX
    Explanations

    written confirmation

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

    The neuron selectively activates on numeric values—especially multi‐digit and decimal numbers representing quantitative data (e.g. probabilities, measures, statistics).

    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
    mehr
    0.96
    <0x0C>
    0.84
     ممکن
    0.82
    Analyzing
    0.79
    ان
    0.78
     impresora
    0.76
     analyze
    0.75
    traverse
    0.75
     alcance
    0.74
    Bar
    0.73
    POSITIVE LOGITS
     confirms
    1.05
     confirmado
    1.04
     Confirm
    0.99
    д
    0.96
     confirming
    0.95
     définitive
    0.94
    ри
    0.93
     definitivamente
    0.93
    ли
    0.92
     potwier
    0.90
    Activations Density 0.249%

    No Known Activations