© 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. 4423
    Prev
    Next
    INDEX
    Explanations

    covariance Matrix Analysis

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

    This neuron detects floating‐point numeric tokens (numbers containing a decimal point).

    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
     柄
    0.88
     取っ
    0.80
    ޖ
    0.79
    㳯
    0.73
    氻
    0.71
    ání
    0.71
     adjudication
    0.71
     Anf
    0.71
    ⌇
    0.71
    ocurrency
    0.70
    POSITIVE LOGITS
    ts
    0.87
    पढ़
    0.84
    et
    0.77
    ted
    0.76
    सीएस
    0.75
    こと
    0.74
    说明
    0.73
    mont
    0.73
    ৬০
    0.71
     hộp
    0.70
    Activations Density 0.002%

    No Known Activations