© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14594
    Prev
    Next
    INDEX
    Explanations

    1, 2, 9 sequence

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

    The main thing this neuron does is find numeric expressions and quantitative references (e.g. digits, years, percentages, ratings, model 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
     Сасик
    1.04
    ঞ্জার
    0.98
    <unused85>
    0.95
    <unused38>
    0.93
    <unused56>
    0.92
     Yatha
    0.91
     Unpublished
    0.91
    नामेंट
    0.89
    <unused475>
    0.89
    <unused411>
    0.88
    POSITIVE LOGITS
    0.88
       
    0.74
    <h3>
    0.72
    0.72
    :
    0.69
     billions
    0.69
     famously
    0.68
         
    0.63
     adalah
    0.63
    ♏
    0.61
    Activations Density 6.151%

    No Known Activations