© 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. 24-GEMMASCOPE-2-RES-16K
    7. 4931
    Prev
    Next
    INDEX
    Explanations

    bed / in:agg

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

    The neuron strongly activates on numerical tokens and numeric expressions (e.g. digits, measurements, version 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
    த
    0.82
    నే
    0.79
    ដើម្បី
    0.77
    evin
    0.75
     дія
    0.74
    纮
    0.72
    ક્ષ
    0.71
    বল
    0.70
     функ
    0.70
    سر
    0.68
    POSITIVE LOGITS
    レクトリ
    0.88
    住了
    0.81
     anxiously
    0.79
     ediciones
    0.79
     Executes
    0.78
    到时候
    0.77
     בכל
    0.76
    jera
    0.75
    ﺘ
    0.75
     gratefully
    0.74
    Activations Density 0.001%

    No Known Activations