© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129468
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on date and time expressions (e.g. month names and adjoining numerals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    month and day names

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     on
    -1.68
     several
    -1.55
     four
    -1.54
     only
    -1.49
     any
    -1.43
     their
    -1.43
     five
    -1.43
     two
    -1.42
     three
    -1.41
     for
    -1.41
    POSITIVE LOGITS
     gesti
    1.23
     applau
    1.20
     BOOM
    1.19
     Dinas
    1.17
     [''],
    1.16
     hemd
    1.14
     kangaroo
    1.09
     gelo
    1.09
     IMDG
    1.09
    don
    1.08
    Activations Density 0.107%

    No Known Activations