© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 19342
    Prev
    Next
    INDEX
    Explanations

    events following an action

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

    The neuron is detecting temporal sequence markers, especially the word “after” (and similar transitional cues like “then” or “when”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    still
    -1.63
     ajouter
    -1.48
     extrait
    -1.47
     autorisation
    -1.47
    masing
    -1.45
     odeur
    -1.43
     Responsable
    -1.41
      
    -1.40
    probably
    -1.39
     chinois
    -1.38
    POSITIVE LOGITS
     with
    2.75
     after
    2.48
     when
    1.87
     that
    1.81
     from
    1.75
     two
    1.66
     of
    1.66
     one
    1.63
     or
    1.46
     despite
    1.44
    Activations Density 0.107%

    No Known Activations