© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6122
    Prev
    Next
    INDEX
    Explanations

    summary description five words

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

    The neuron strongly activates on numeric tokens (especially long or decimal 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
    उनकी
    0.75
    ில்
    0.74
    <0x0D>
    0.71
    og
    0.70
    詆
    0.70
     annuel
    0.69
    他们的
    0.68
     serius
    0.68
    ق
    0.67
    queue
    0.66
    POSITIVE LOGITS
    ند
    0.80
     tämä
    0.79
    ட்சத்திர
    0.79
     zastos
    0.77
    มะ
    0.77
     Está
    0.76
     واذا
    0.76
    nél
    0.74
     اور
    0.73
     attaining
    0.72
    Activations Density 0.019%

    No Known Activations