© 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. 22-GEMMASCOPE-RES-131K
    7. 93191
    Prev
    Next
    INDEX
    Explanations

    code and script headers

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

    The neuron fires on tokens containing non-letter punctuation (e.g. apostrophes ’s, hyphens so-, or digits/decimal points).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    znym
    -1.05
     évi
    -0.93
    شرة
    -0.91
    花火
    -0.91
     while
    -0.90
     poetas
    -0.90
    блема
    -0.88
     from
    -0.87
     conseguido
    -0.86
     emel
    -0.86
    POSITIVE LOGITS
     価格
    0.99
     purchases
    0.83
    分鐘
    0.82
     որ
    0.82
     vores
    0.82
     eröffnet
    0.81
     season
    0.80
     Chopper
    0.80
     probs
    0.80
     restaurant
    0.79
    Activations Density 0.000%

    No Known Activations