© 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. 8554
    Prev
    Next
    INDEX
    Explanations

    numbers following 9

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

    The neuron activates on standalone multi‐digit numeric tokens, especially four-digit numbers such as publication years or reference identifiers.

    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
    it
    0.56
     except
    0.56
    '',
    0.54
    '$\
    0.53
    ]}\
    0.52
    ";
    0.52
    を除
    0.52
    itani
    0.51
     Bacteriol
    0.51
    ↵↵↵
    0.50
    POSITIVE LOGITS
    ތ
    0.65
     மதுரை
    0.64
     legnagyobb
    0.63
     respons
    0.63
     Daarom
    0.63
     Podemos
    0.62
     partido
    0.61
    モンド
    0.60
     اہم
    0.60
     problèmes
    0.60
    Activations Density 0.039%

    No Known Activations