© 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. 34-GEMMASCOPE-RES-131K
    7. 99830
    Prev
    Next
    INDEX
    Explanations

    after a [quantity/duration]

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

    The neuron fires on temporal phrases that begin “after a…” followed by a duration (e.g. “after a couple of years,” “after a short stay,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     before
    -1.75
     now
    -1.44
     after
    -1.32
     also
    -1.21
     new
    -1.16
    ilité
    -1.10
     as
    -1.07
    と思ったら
    -1.04
     Before
    -1.04
     actually
    -1.03
    POSITIVE LOGITS
     initial
    1.63
     failed
    1.55
     unsuccessful
    1.49
    Initial
    1.48
    initial
    1.34
     successful
    1.25
     awhile
    1.16
     مباشرة
    1.16
     Initial
    1.14
     unsuccessfully
    1.14
    Activations Density 0.112%

    No Known Activations