© 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. 65979
    Prev
    Next
    INDEX
    Explanations

    times after at

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

    This neuron detects the preposition “at” when it introduces an event’s start time.

    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
     not
    -1.52
     how
    -1.44
     =
    -1.33
     -
    -1.32
     that
    -1.25
     potential
    -1.21
     other
    -1.20
     then
    -1.20
    how
    -1.20
    not
    -1.16
    POSITIVE LOGITS
     великолеп
    1.55
     noon
    1.47
     tevé
    1.41
     feroit
    1.30
     rumores
    1.27
     turquesa
    1.25
    öffnung
    1.25
     शुर
    1.23
     gebruikt
    1.23
     благода
    1.23
    Activations Density 0.015%

    No Known Activations