© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14274
    Prev
    Next
    INDEX
    Explanations

    time of day (am/pm)

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

    The neuron strongly activates on numeric tokens that are part of date/time stamps (e.g. hours, minutes, and AM/PM markers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cualquier
    -1.59
     Consulte
    -1.41
     Coloque
    -1.17
     Pueden
    -1.16
     cualquier
    -1.13
     recoge
    -1.10
     любую
    -1.10
     negli
    -1.09
     Vorteil
    -1.08
     Tienes
    -1.08
    POSITIVE LOGITS
     поэтому
    1.18
    ianie
    0.99
     они
    0.98
     crevices
    0.98
     зима
    0.95
     seuil
    0.95
     Zwar
    0.94
     jotka
    0.92
     ultra
    0.91
     вони
    0.91
    Activations Density 0.027%

    No Known Activations