© 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. 15420
    Prev
    Next
    INDEX
    Explanations

    past experiences and specific times

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

    The neuron detects first‐person past‐tense narrative segments—i.e., “I” plus past‐tense verbs recounting personal experiences.

    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
     currently
    -2.25
     yesterday
    -2.16
     today
    -2.02
    目前
    -1.91
    去年
    -1.82
     Yesterday
    -1.59
     aujourd
    -1.54
     now
    -1.51
     recently
    -1.50
     현재
    -1.50
    POSITIVE LOGITS
     was
    1.86
     тогда
    1.77
     damals
    1.73
     hongos
    1.64
     my
    1.40
    当時の
    1.40
     would
    1.34
     memorable
    1.33
     IIRC
    1.32
     damaligen
    1.30
    Activations Density 0.050%

    No Known Activations