© 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. 5489
    Prev
    Next
    INDEX
    Explanations

    variants, combinations, different

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

    The neuron primarily fires on technical acronyms, measurement codes, and other all-caps abbreviations used in clinical or research study write-ups.

    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
    Businesses
    0.98
    businesses
    0.95
    requestId
    0.94
    welcome
    0.90
    かなり
    0.90
    誰
    0.90
    ธุรกิจ
    0.89
    financ
    0.89
    newsletter
    0.88
    Instit
    0.88
    POSITIVE LOGITS
     simulated
    0.99
     versus
    0.96
     $+
    0.94
     centrifuged
    0.92
     inoculated
    0.91
     variants
    0.91
     AcOH
    0.90
     combinations
    0.89
     tested
    0.89
     adsorbed
    0.88
    Activations Density 0.284%

    No Known Activations