© 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. 31-GEMMASCOPE-2-RES-16K
    7. 15624
    Prev
    Next
    INDEX
    Explanations

    another example or method

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

    The neuron activates on tokens that are part of numeric references—especially document or standard identification numbers with decimal points.

    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
    options
    0.67
    statements
    0.64
     statements
    0.61
    ultz
    0.60
    回事
    0.60
     कथन
    0.58
    ?;
    0.58
     pkt
    0.57
    itive
    0.55
    archar
    0.55
    POSITIVE LOGITS
     パー
    0.62
     центра
    0.59
     Persever
    0.59
     Prohibition
    0.57
     soapy
    0.56
     CSF
    0.54
     风
    0.54
     Decentral
    0.54
     prohibited
    0.53
     barrel
    0.53
    Activations Density 0.000%

    No Known Activations