© 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. 42728
    Prev
    Next
    INDEX
    Explanations

    career transitions

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

    The neuron is acting as a detector for relatively rare or content‐bearing subword tokens—that is, uncommon vocabulary pieces rather than high-frequency function 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_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
    llavero
    -1.39
     اخیر
    -1.32
     errorMessage
    -1.24
     loue
    -1.23
    también
    -1.23
    ILERS
    -1.21
    corazón
    -1.19
    dagogik
    -1.19
     läsa
    -1.16
     cesse
    -1.15
    POSITIVE LOGITS
    它
    1.56
     it
    1.42
     in
    1.38
    helial
    1.34
    GOTREF
    1.34
     dissatisfied
    1.27
     nếu
    1.27
    romas
    1.26
     ainda
    1.23
    !("
    1.22
    Activations Density 0.166%

    No Known Activations