© 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. 28979
    Prev
    Next
    INDEX
    Explanations

    alternator rectifier

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

    The neuron specializes in spotting technical specification snippets—particularly numbers (and number–unit pairs) and related mechanical/electrical component terms.

    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
    րան
    -1.42
     Cocktails
    -1.26
    ꯰
    -1.16
     mutiara
    -1.14
    スタム
    -1.13
    そんな
    -1.13
    そこで
    -1.13
    ⎛
    -1.13
     樹
    -1.12
     Hibiscus
    -1.12
    POSITIVE LOGITS
     not
    1.42
     Many
    1.27
     molte
    1.21
    .
    1.16
    zości
    1.16
     up
    1.14
     for
    1.13
     u
    1.05
     my
    1.05
     Ignoring
    1.04
    Activations Density 0.032%

    No Known Activations