© 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. 34-GEMMASCOPE-RES-131K
    7. 66253
    Prev
    Next
    INDEX
    Explanations

    events follow or come

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

    This neuron fires on common sentence-initial or lead-in words (e.g. “Today,” “It,” “The,” etc.), effectively marking the beginnings of news‐style sentences.

    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
    remains
    -1.00
    }%
    -0.96
    ()])
    -0.96
    ización
    -0.94
    豔
    -0.92
    isc
    -0.90
    remain
    -0.89
    vergleich
    -0.89
    tienen
    -0.88
    😉
    -0.88
    POSITIVE LOGITS
     comes
    4.19
     follows
    4.09
     follow
    2.89
     come
    2.73
     Follows
    2.47
     Comes
    2.45
    follows
    2.41
    comes
    2.41
    Comes
    2.20
     folgt
    2.17
    Activations Density 0.080%

    No Known Activations