© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 90683
    Prev
    Next
    INDEX
    Explanations

    multilingual transitions after concepts

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

    The neuron fires strongly on business‐and‐technology buzzwords—especially nouns and metrics describing product features, capabilities, or performance.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     but
    -1.97
     if
    -1.77
     although
    -1.70
     after
    -1.50
     mainly
    -1.39
     having
    -1.36
     because
    -1.35
     mostly
    -1.34
     or
    -1.26
     obviously
    -1.24
    POSITIVE LOGITS
     to
    1.91
     mentre
    1.74
     across
    1.63
     mientras
    1.54
     WHILE
    1.50
     enquanto
    1.47
    เพื่อ
    1.43
     while
    1.40
    while
    1.38
     grâce
    1.33
    Activations Density 0.083%

    No Known Activations