© 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. 28220
    Prev
    Next
    INDEX
    Explanations

    technical code and text

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

    The neuron activates on sequences of digits in code (numeric literals or numbers embedded in identifiers).

    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
     when
    -1.67
     every
    -1.56
     intricate
    -1.55
     exquisite
    -1.52
     eleven
    -1.51
     before
    -1.42
     three
    -1.40
     more
    -1.37
     first
    -1.35
     in
    -1.30
    POSITIVE LOGITS
     ingenieros
    1.54
     nieuwe
    1.51
     ridiculously
    1.38
    1.37
    졍
    1.32
     utilisée
    1.31
    这本书
    1.30
     professione
    1.30
    新しく
    1.30
     puntuación
    1.30
    Activations Density 0.001%
    No Known Activations