© 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. 31-GEMMASCOPE-2-RES-16K
    7. 11531
    Prev
    Next
    INDEX
    Explanations

    list separators and continuations

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

    The neuron is looking for numeric tokens—especially floating-point or other numerical measurements and 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_31_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
    ishman
    0.69
    á
    0.66
    ư
    0.65
    ád
    0.64
    ú
    0.64
    mysql
    0.61
    orent
    0.61
    ｣
    0.61
    ál
    0.61
    oron
    0.60
    POSITIVE LOGITS
     others
    0.94
     eventuali
    0.83
     etc
    0.81
     various
    0.80
     putative
    0.80
     Villiers
    0.77
     proprietary
    0.76
     wobei
    0.75
     unnamed
    0.75
     undisclosed
    0.75
    Activations Density 0.393%

    No Known Activations