© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 4886
    Prev
    Next
    INDEX
    Explanations

    fine-tuned

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

    The neuron fires on explicit quantity indicators—cardinal numbers and quantifying words (e.g. “one,” “few,” numeric references).

    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
     DataFrame
    0.46
    ^{'
    0.46
     signified
    0.46
     ①
    0.46
     monadic
    0.46
    This
    0.45
     monoton
    0.44
     inputted
    0.43
     {
    0.43
     associativity
    0.43
    POSITIVE LOGITS
     biofuels
    0.52
     катастро
    0.51
    ﯿ
    0.48
     крупней
    0.48
     गेम्स
    0.47
     fledgling
    0.46
     घोटाले
    0.46
    quinox
    0.46
    islang
    0.46
     fisheries
    0.46
    Activations Density 1.323%

    No Known Activations