© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6499
    Prev
    Next
    INDEX
    Explanations

    19xx dates

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

    The neuron activates on numeric tokens (particularly years and day/month numbers) in date formats, especially those adjacent to slashes.

    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
    ૯
    0.78
    ۹
    0.71
    ９
    0.68
    契
    0.67
    ୯
    0.66
    ೯
    0.65
    ٩
    0.64
     ९
    0.63
     recientemente
    0.61
     hatta
    0.60
    POSITIVE LOGITS
    শ্বর
    0.57
    ాలని
    0.57
    carboxylic
    0.55
     Pandemic
    0.55
    BOR
    0.54
     BOOL
    0.54
    ើង
    0.54
     হাফ
    0.53
     mumbled
    0.53
    ebilir
    0.53
    Activations Density 0.078%

    No Known Activations