© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35314
    Prev
    Next
    INDEX
    Explanations

    social media usage

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

    The neuron responds to clause‐ or sentence‐introducing function words or discourse markers (e.g. “If,” “This,” “Just,” “it” at the start of clauses).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     pedagogy
    -0.90
     tolua
    -0.88
    lenburg
    -0.87
    hores
    -0.84
    ednes
    -0.84
    directly
    -0.84
     pedagog
    -0.84
     capitalismo
    -0.82
    смарт
    -0.82
     excep
    -0.81
    POSITIVE LOGITS
    Académie
    0.94
     how
    0.93
    washington
    0.93
    There
    0.93
     cilvē
    0.91
    timbangkan
    0.91
    Bacterial
    0.90
     your
    0.88
     ਹੋ
    0.87
    How
    0.87
    Activations Density 0.001%

    No Known Activations