© 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. 34-GEMMASCOPE-RES-131K
    7. 48414
    Prev
    Next
    INDEX
    Explanations

    week in, week of

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

    The neuron looks for mentions of calendar weeks or “week of” date references, i.e. the word “week” (often capitalized) paired with numbers or “of.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     today
    -1.16
     yesterday
    -1.15
     reducido
    -1.13
     seconds
    -1.13
     rewarding
    -1.13
     mittler
    -1.11
     revamped
    -1.10
     vastly
    -1.09
     locales
    -1.08
     doros
    -1.08
    POSITIVE LOGITS
    end
    1.63
    long
    1.48
    enden
    1.43
    week
    1.42
    ender
    1.41
    lies
    1.32
    ends
    1.27
    night
    1.22
    л
    1.20
    ☔
    1.17
    Activations Density 0.038%

    No Known Activations