© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7994
    Prev
    Next
    INDEX
    Explanations

    list formatting

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

    The neuron strongly activates on numeric values and their associated quantitative measurements (e.g., coordinates, voltages, counts, contact 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_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
    ד
    0.81
    ח
    0.78
     ח
    0.75
    ح
    0.74
     remainder
    0.70
     CHARACTER
    0.67
    roku
    0.65
    tod
    0.64
     Accessed
    0.63
    ๊
    0.63
    POSITIVE LOGITS
    ма
    0.84
     Saheb
    0.78
    jours
    0.77
    ↺
    0.75
    ्यात
    0.73
     Quatre
    0.73
     Düsseldorf
    0.72
     abbey
    0.72
    アルバム
    0.71
    ком
    0.71
    Activations Density 0.000%

    No Known Activations