© 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. 4720
    Prev
    Next
    INDEX
    Explanations

    DSL code snippets, financial terms, source information

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

    The neuron strongly activates on numeric tokens (e.g. digits, percentages, decimal values), i.e. it detects numbers.

    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
     Atmosphere
    0.80
     tradizione
    0.79
     žád
    0.75
    아
    0.75
     seashore
    0.74
    污染物
    0.73
     episodio
    0.72
     वापस
    0.71
     cómo
    0.71
     teniendo
    0.71
    POSITIVE LOGITS
    Xinhua
    0.77
    摆
    0.73
    ujjati
    0.71
    啮
    0.71
    {``
    0.70
    ları
    0.70
    eyiz
    0.68
    }\{\
    0.68
     hinges
    0.68
    复制代码
    0.67
    Activations Density 0.000%

    No Known Activations