© 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. 70705
    Prev
    Next
    INDEX
    Explanations

    explaining observed phenomena

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

    The neuron selectively activates on verbs that introduce explanations or descriptions (e.g. “explain,” “explains,” “describe”).

    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
     a
    -1.87
     you
    -1.81
     tyto
    -1.61
     esclavos
    -1.59
     però
    -1.55
     preocupado
    -1.55
     versátil
    -1.49
    इटम
    -1.49
     participado
    -1.49
     può
    -1.48
    POSITIVE LOGITS
    </em>
    1.59
     why
    1.57
    .”
    1.53
     which
    1.52
    那些
    1.35
    }{
    1.30
    

    1.30
     what
    1.27
    <em>
    1.24
     in
    1.24
    Activations Density 0.056%

    No Known Activations