© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 68540
    Prev
    Next
    INDEX
    Explanations

    present day description

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

    The neuron reliably fires on the token “Today” (and its immediate context) as a marker that the text is shifting to a present‐day status or update.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     first
    -1.27
     from
    -1.26
     while
    -1.26
     mãi
    -1.26
     during
    -1.19
     причем
    -1.18
     will
    -1.18
     or
    -1.14
     nunca
    -1.14
     came
    -1.13
    POSITIVE LOGITS
    ····
    1.27
     fendi
    1.23
    currently
    1.20
     rayures
    1.16
     الآن
    1.13
    Nowadays
    1.13
    still
    1.12
     corriger
    1.10
     fabrik
    1.09
     étages
    1.09
    Activations Density 0.029%

    No Known Activations