© 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. 5975
    Prev
    Next
    INDEX
    Explanations

    code and symbols

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

    The neuron activates primarily on numeric tokens representing measured values or decimal numbers (e.g. “375.7666,” “642.7147,” etc.).

    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
    TN
    0.50
    0
    0.50
    /
    0.50
    view
    0.50
    ('\\
    0.49
     […]
    0.49
    4
    0.48
    et
    0.48
    9
    0.48
    a
    0.47
    POSITIVE LOGITS
     astfel
    0.63
    គឺជា
    0.61
     هیڅ
    0.59
     इतर
    0.54
     તરીકે
    0.54
     위해
    0.53
     करि
    0.53
    㶪
    0.53
     दोन्ही
    0.52
     fiind
    0.52
    Activations Density 0.002%

    No Known Activations