© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2353
    Prev
    Next
    INDEX
    Explanations

    example of specific

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

    This neuron strongly activates on numeric expressions (especially decimal-formatted numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    punctuation marks, especially commas and periods.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    high-frequency function words in running text (articles, auxiliaries, and pronouns).

    oai_token-act-pair · gpt-5Triggered 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
    מש
    0.71
    ich
    0.71
    odob
    0.71
    сила
    0.70
    ர
    0.68
    CD
    0.68
    الصفحه
    0.67
    ृ
    0.66
    erm
    0.66
    телно
    0.66
    POSITIVE LOGITS
     eficiencia
    0.82
     Leone
    0.80
     BEE
    0.78
     bleibt
    0.77
     terão
    0.77
     EFF
    0.77
    squa
    0.77
     পরই
    0.75
     Humboldt
    0.74
     permanec
    0.74
    Activations Density 0.000%

    No Known Activations