© 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. 7606
    Prev
    Next
    INDEX
    Explanations

    historical time

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

    The neuron strongly activates on the word “time” (and its immediate context), effectively detecting mentions of “time.”

    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
    𓄹
    -1.91
     possibility
    -1.91
     following
    -1.86
     clínico
    -1.84
    にならない
    -1.77
    
    -1.77
    ⡪
    -1.68
    〜。
    -1.68
    خة
    -1.66
     Möglichkeit
    -1.63
    POSITIVE LOGITS
    </h2>
    1.98
    v
    1.61
    x
    1.59
    n
    1.58
    </
    1.56
     exacting
    1.51
    ",
    1.48
    ut
    1.47
    ta
    1.47
    (
    1.43
    Activations Density 0.007%

    No Known Activations