© 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. 98316
    Prev
    Next
    INDEX
    Explanations

    in the past

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

    The main thing this neuron does is detect past-tense narrative markers—verbs and key words that frame descriptions of events or recollections.

    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
     aujourd
    -1.23
    についても
    -1.11
     today
    -1.09
     薬局
    -1.09
    がありますが
    -1.05
     atualização
    -1.05
    sometimes
    -1.04
    については
    -1.04
     никогда
    -1.04
     now
    -1.02
    POSITIVE LOGITS
     ещё
    1.10
     other
    1.10
     еще
    1.09
    他に
    1.07
     still
    1.06
     вокруг
    1.05
     только
    1.05
    ☎️
    1.03
     ohne
    1.02
    >";
    1.02
    Activations Density 0.044%

    No Known Activations