© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10863
    Prev
    Next
    INDEX
    Explanations

    Millennium Prize Problem, School Year

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

    This neuron fires on tokens immediately preceding large numeric identifiers (e.g. DOIs, PMIDs, database or citation numbers).

    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
    Р
    0.87
    Т
    0.86
    Ф
    0.84
    Б
    0.82
    О
    0.81
    Когда
    0.80
    #!/
    0.80
    Л
    0.79
    Ди
    0.76
    Ги
    0.76
    POSITIVE LOGITS
    ằm
    0.84
    ach
    0.82
     mitochondria
    0.75
     Seeger
    0.75
    ia
    0.73
    umoto
    0.73
    ann
    0.72
    ai
    0.72
    um
    0.71
    lių
    0.71
    Activations Density 0.000%

    No Known Activations