© 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. 81206
    Prev
    Next
    INDEX
    Explanations

    legal, scientific, or navigational terms

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

    The neuron is keyed to single standout nouns used as formal section or heading labels (e.g. “Part,” “Charge,” “Land,” “Arrest,” “Threatened”) rather than to ordinary running-text words.

    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
     which
    -2.50
     including
    -1.27
     where
    -1.22
     it
    -1.21
     for
    -1.13
    which
    -1.09
     Which
    -1.05
     tiež
    -1.00
     incluyendo
    -0.98
     this
    -0.95
    POSITIVE LOGITS
    χος
    1.09
    mentar
    1.05
     /><
    0.98
     persecu
    0.96
    omatous
    0.96
    ्ञ
    0.95
     chỉ
    0.95
     natale
    0.94
     arac
    0.94
     sincron
    0.94
    Activations Density 0.003%

    No Known Activations