© 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. 10-GEMMASCOPE-RES-131K
    7. 112386
    Prev
    Next
    INDEX
    Explanations

    four year durations

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

    The neuron detects mentions of time spans expressed in years (e.g. “four-year” or “four years”).

    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
    ↵
    -2.22
    sandalia
    -2.19
     中身
    -2.09
    geſ
    -2.09
    харш
    -2.05
    镡
    -2.03
     destes
    -2.02
    蚡
    -1.96
    灃
    -1.89
    </b>
    -1.84
    POSITIVE LOGITS
    4
    2.48
     of
    2.48
    x
    2.31
    萵
    2.22
    ise
    2.19
    煳
    2.19
    r
    2.13
    For
    2.11
    媖
    2.09
    The
    2.05
    Activations Density 0.024%

    No Known Activations