© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 127367
    Prev
    Next
    INDEX
    Explanations

    HTML transitional and code comments

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

    The neuron fires on sentence‐level structural markers—capitalized words that begin sentences (like “This,” “If,” “When”) and the punctuation that ends them.

    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
    XXL
    -1.55
    !(
    -1.51
    鈇
    -1.41
     adam
    -1.41
    iree
    -1.34
    钾
    -1.33
    mout
    -1.30
    yip
    -1.30
    !(
    -1.28
    chance
    -1.27
    POSITIVE LOGITS
     here
    1.93
     then
    1.68
     We
    1.46
    \
    1.41
     years
    1.41
    jenige
    1.41
    year
    1.39
    についても
    1.37
     finding
    1.37
     programs
    1.37
    Activations Density 0.002%

    No Known Activations