© 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. 34-GEMMASCOPE-RES-131K
    7. 119356
    Prev
    Next
    INDEX
    Explanations

    syntactic, synthesis, consecutive, continuously

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

    The neuron is highly activated by the Greek letter sequence “συ” (as in συζήτηση, Συμβούλιο, κ.λπ.) and by the Latin‐script prefix “syn,” so it’s detecting the prefix “syn/συν.”

    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
     ayudan
    -0.84
     Villas
    -0.81
    重量
    -0.81
     DANIEL
    -0.78
     skupiny
    -0.76
     resultan
    -0.76
     точку
    -0.75
     architectures
    -0.74
    welling
    -0.74
    separated
    -0.74
    POSITIVE LOGITS
     kılıf
    0.82
     consecutive
    0.81
     Manusia
    0.79
     зви
    0.75
     continuous
    0.74
    Usually
    0.74
    Contributing
    0.73
    compan
    0.73
     }^{*}$
    0.73
    つづく
    0.71
    Activations Density 0.008%

    No Known Activations