© 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. 34-GEMMASCOPE-RES-131K
    7. 107255
    Prev
    Next
    INDEX
    Explanations

    needle and thread

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

    The main thing this neuron does is detect references to JSON operations (e.g. JSON.parse or JSON.stringify).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ルター
    -0.73
    layan
    -0.71
     Xin
    -0.69
    uland
    -0.69
    hoek
    -0.68
    ความ
    -0.66
    Anna
    -0.65
    𝖎
    -0.64
    xFFFF
    -0.64
    EG
    -0.64
    POSITIVE LOGITS
     string
    3.30
     strung
    3.08
     threading
    2.73
    string
    2.67
     thread
    2.42
     Thread
    2.39
     String
    2.36
     strings
    2.25
     threaded
    2.20
    thread
    2.19
    Activations Density 0.035%

    No Known Activations