© 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. 61431
    Prev
    Next
    INDEX
    Explanations

    part of

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

    The neuron detects instances of “part,” especially in “part of” phrases indicating membership or inclusion in a group or program.

    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
     zufried
    -1.85
    gemeester
    -1.76
    гряз
    -1.71
     einwand
    -1.61
     time
    -1.59
    iyi
    -1.55
     alojamientos
    -1.53
    颔
    -1.53
     they
    -1.50
    祧
    -1.48
    POSITIVE LOGITS
    Джере
    1.68
    笮
    1.66
    Ито
    1.63
    ou
    1.57
    Примеча
    1.54
     are
    1.52
     chers
    1.40
     виправи
    1.39
    Условия
    1.38
    </b>
    1.37
    Activations Density 0.030%

    No Known Activations