© 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. 68809
    Prev
    Next
    INDEX
    Explanations

    thinking about

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

    This neuron activates on occurrences of “think” (and its morphological variants like thinking, thought(s), Think).

    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
     momentan
    -2.31
     フリル
    -1.99
    关系
    -1.99
    I
    -1.97
    Elaboración
    -1.96
    澌
    -1.91
    Kommentare
    -1.91
    ⃙
    -1.88
     описа
    -1.84
    ?’
    -1.83
    POSITIVE LOGITS
    {
    2.44
    .
    2.38
     about
    2.08
     think
    2.02
    ykjav
    1.99
    ológicos
    1.88
     fondly
    1.81
     intently
    1.77
    ↵↵
    1.75
    Thoughts
    1.70
    Activations Density 0.090%

    No Known Activations