© 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. 59404
    Prev
    Next
    INDEX
    Explanations

    first time, last time

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

    This neuron responds to occurrences of the word “time” in temporal expressions (e.g. “the last time,” “the first 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
    )$.
    -2.00
     become
    -1.92
     more
    -1.91
        
    -1.89
    อีก
    -1.86
     and
    -1.81
    jenigen
    -1.80
     at
    -1.78
     tráfico
    -1.71
     With
    -1.69
    POSITIVE LOGITS
     you
    3.94
     we
    3.11
     they
    3.05
     it
    2.30
     your
    2.08
     familières
    2.08
     enjeux
    2.00
     chrétiens
    1.98
     convain
    1.96
     rigue
    1.94
    Activations Density 0.072%

    No Known Activations