© 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. 13806
    Prev
    Next
    INDEX
    Explanations

    best friend, prime time, today, strength hands

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

    The neuron fires on numeric tokens—i.e. any string of digits (including years, docket or section numbers, and decimal‐formatted 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
    ר
    0.94
    ли
    0.92
    се
    0.88
    ش
    0.87
    ста
    0.81
    <0xA7>
    0.80
    ч
    0.80
    м
    0.80
    лу
    0.79
    дного
    0.76
    POSITIVE LOGITS
     confiscated
    0.90
     Susanne
    0.83
     unoccupied
    0.80
     unexplored
    0.80
     suspe
    0.77
    现状
    0.77
     plagued
    0.76
     subsequ
    0.75
     cryptic
    0.75
     prohibited
    0.74
    Activations Density 0.000%

    No Known Activations