© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130303
    Prev
    Next
    INDEX
    Explanations

    This neuron specifically activates on the past-tense verb “thought.”

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

    thought

    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
    釆
    -1.51
    他們
    -1.45
     other
    -1.42
     Diputados
    -1.42
     כאשר
    -1.35
     sebagainya
    -1.34
     In
    -1.33
     whatnot
    -1.30
    ↵↵↵↵↵↵↵
    -1.29
     which
    -1.28
    POSITIVE LOGITS
     it
    2.20
     that
    2.11
    {
    1.66
    筇
    1.45
     they
    1.33
     to
    1.31
     geeigne
    1.28
     {}));
    1.27
    …」
    1.26
    [];
    
    1.26
    Activations Density 0.033%

    No Known Activations