© 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. 36136
    Prev
    Next
    INDEX
    Explanations

    network requests and json data

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

    The neuron is detecting numeric tokens—literal numbers (digits, numeric constants, counts, status codes, etc.) in the code or text.

    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
    ǣ
    -1.00
    efore
    -0.95
     céle
    -0.94
    tiem
    -0.94
    efined
    -0.93
    voller
    -0.93
     виправи
    -0.89
    сар
    -0.89
     mieste
    -0.89
    trek
    -0.87
    POSITIVE LOGITS
     request
    1.31
     json
    1.28
     их
    1.13
    get
    1.06
     requested
    1.05
     requests
    1.04
     JSON
    1.03
    json
    1.02
     needed
    1.01
     respond
    0.96
    Activations Density 0.022%

    No Known Activations