© 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. 10-GEMMASCOPE-RES-131K
    7. 19060
    Prev
    Next
    INDEX
    Explanations

    going

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

    The neuron exclusively activates on occurrences of the word “going.”

    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
    “
    -2.80
    {
    -2.30
     give
    -2.23
     eléctricas
    -2.16
    ным
    -2.09
     not
    -2.08
    ین
    -1.98
     seems
    -1.97
    ให้
    -1.97
     emocionante
    -1.96
    POSITIVE LOGITS
    戋
    2.56
     They
    2.55
    摭
    2.38
     validamos
    2.31
     employing
    2.30
    熥
    2.27
    弍
    2.22
     Importantly
    2.22
     Specifically
    2.20
     hierzu
    2.17
    Activations Density 0.009%

    No Known Activations