© 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. 5376
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric expressions and comparative terms (e.g. version numbers, measurements, ages like “older,” “later,” or any token containing digits).

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

    the word "of" when it appears in formal or technical contexts, particularly in phrases indicating relationships or specifications.

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

    connective and modifier terms that signal addition, alternatives, comparisons, or degree within a sentence.

    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
     horaires
    0.77
     Herbs
    0.76
     প্রদেশ
    0.70
     caul
    0.70
    closeup
    0.70
     Pilots
    0.68
     Posters
    0.68
     Answers
    0.66
     hồ
    0.66
     Côte
    0.66
    POSITIVE LOGITS
    サート
    0.87
     сочета
    0.83
     создание
    0.82
     школы
    0.82
    となった
    0.81
    те
    0.78
     окружающей
    0.77
     супру
    0.77
     мысли
    0.76
     специальных
    0.75
    Activations Density 0.000%

    No Known Activations