© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71247
    Prev
    Next
    INDEX
    Explanations

    previous roles and positions

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

    This neuron detects temporal transition markers in professional biographies—words like “Prior,” “Previously,” “served,” and “was” that introduce past roles or career steps.

    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
    kkor
    -1.48
    Controllo
    -1.47
    によっては
    -1.41
    Efter
    -1.41
    chọn
    -1.38
    therosclerosis
    -1.35
    Från
    -1.34
    tlere
    -1.34
    rakt
    -1.30
    مقال
    -1.29
    POSITIVE LOGITS
     in
    1.81
    previously
    1.35
     previously
    1.34
    \
    1.30
     but
    1.30
    .
    1.28
    :
    1.26
    らず
    1.23
    S
    1.20
    --
    1.19
    Activations Density 0.063%

    No Known Activations