© 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. 131059
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on occurrences of the word “set” (particularly in uppercase or code‐related contexts such as macro names and commands).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    setting configurations or values

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    w
    -1.70
    랬
    -1.68
    几个人
    -1.59
    茕
    -1.54
    がありますが
    -1.51
    ፳
    -1.48
    aware
    -1.47
     вони
    -1.46
    これは
    -1.45
    である
    -1.43
    POSITIVE LOGITS
     the
    2.14
     and
    2.13
     their
    1.86
     a
    1.68
    ஂ
    1.63
     as
    1.60
     &
    1.58
    </i>
    1.52
     an
    1.46
    喼
    1.42
    Activations Density 0.030%

    No Known Activations