© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7791
    Prev
    Next
    INDEX
    Explanations

    source or final answer

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

    The neuron fires on unusual floating‐point–style numbers (decimal strings), effectively spotting numeric tokens with a fractional component.

    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_31_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
     […]
    0.57
     మాట్లాడుతూ
    0.55
    validate
    0.53
    […]
    0.52
     réellement
    0.52
     मुताबिक
    0.48
     [...]
    0.47
    caret
    0.47
    sufficiency
    0.47
     hacía
    0.46
    POSITIVE LOGITS
     লেখক
    0.63
     ARTICLE
    0.61
    <i>
    0.59
     Featured
    0.59
     SOURCE
    0.59
     Source
    0.57
    Featured
    0.56
    Correction
    0.56
    Источник
    0.56
     Adapted
    0.55
    Activations Density 0.066%

    No Known Activations