© 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. 12871
    Prev
    Next
    INDEX
    Explanations

    year 2000

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

    The neuron is primarily triggered by stray floating-point number tokens (formatting/artifact numbers) rather than actual words.

    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
    ’
    0.74
     своему
    0.73
    ের
    0.68
    <0x80>
    0.65
    б
    0.64
     svoj
    0.64
    azione
    0.64
    ానని
    0.64
     nerveux
    0.63
    j
    0.63
    POSITIVE LOGITS
     具体
    0.82
    ্টেন
    0.79
     Gandh
    0.77
     Crystallogr
    0.77
    TE
    0.75
    Substituting
    0.75
    MORDOR
    0.73
    具体
    0.71
     Transformations
    0.70
     Özellikle
    0.70
    Activations Density 0.000%

    No Known Activations