© 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. 277
    Prev
    Next
    INDEX
    Explanations

    foreign language words

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

    The neuron activates on tokens that are part of numerical values (digits, multi‐digit numbers, decimals, etc.).

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

    decimal points in numerical values, particularly in financial or statistical data.

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

    numeric figures in text—digits forming quantities, currency amounts, percentages, dates, and measurements, often with commas or decimal points

    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
     idea
    0.17
     time
    0.17
     vaso
    0.16
     moment
    0.15
     tratar
    0.15
     nucleation
    0.15
     scl
    0.15
     uproar
    0.15
     hypo
    0.15
     step
    0.14
    POSITIVE LOGITS
     bốn
    0.18
    5
    0.18
    4
    0.18
    8
    0.18
    7
    0.18
     हजार
    0.17
    ﺎ
    0.17
     総
    0.17
    9
    0.17
    <unused2165>
    0.17
    Activations Density 0.568%

    No Known Activations