© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103496
    Prev
    Next
    INDEX
    Explanations

    kind of

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

    The neuron detects categorical phrases using “kind of” or “type of” to introduce a classification.

    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
    涠
    -2.14
     iden
    -1.83
     emphat
    -1.77
    岨
    -1.77
     starte
    -1.77
    oughby
    -1.73
     uan
    -1.72
    foundland
    -1.68
     disgra
    -1.67
     手順
    -1.66
    POSITIVE LOGITS
     places
    2.03
     way
    1.91
     place
    1.63
    _
    1.60
     approach
    1.59
     time
    1.58
    (
    1.57
    ."
    1.54
     just
    1.54
     possibly
    1.52
    Activations Density 0.009%

    No Known Activations