© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7305
    Prev
    Next
    INDEX
    Explanations

    temporal order and sequence

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

    The neuron is highly sensitive to terms expressing temporal movement or manipulation (e.g. “back,” “time,” “fast,” “forward,” “rewind”)—i.e. references to going back or forward in time.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     हाइड्र
    0.61
     업체
    0.59
     RCB
    0.55
    뿡
    0.54
    オイル
    0.54
     Jardin
    0.54
     ulei
    0.54
     কঠ
    0.53
     केंद्रों
    0.53
    讧
    0.53
    POSITIVE LOGITS
     chronological
    1.84
     timeline
    1.77
     chronology
    1.75
     temporal
    1.74
     retrospective
    1.66
     과거
    1.66
    temporal
    1.63
     retroactive
    1.63
     retrospect
    1.62
    historical
    1.62
    Activations Density 0.814%

    No Known Activations