© 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. 8959
    Prev
    Next
    INDEX
    Explanations

    a / an followed by a word

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

    This neuron detects standalone technical terms or jargon—especially names of processes, commands, functions, or concept headings.

    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
    抧
    0.73
    㓶
    0.70
     Maths
    0.67
    CLUDING
    0.66
     겨울
    0.66
     海外
    0.66
     avanzado
    0.65
     COMMUNICATIONS
    0.64
     Sts
    0.64
    偒
    0.64
    POSITIVE LOGITS
     terroir
    0.86
     stumpage
    0.81
     istilah
    0.79
    cohort
    0.75
     epitopes
    0.75
     hyperparameters
    0.73
     holotype
    0.72
     quorum
    0.71
     anastomosis
    0.70
    daemon
    0.69
    Activations Density 0.609%

    No Known Activations