© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14048
    Prev
    Next
    INDEX
    Explanations

    request or update

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

    The neuron detects code tokens where an HTTP request is being constructed or issued (e.g. “Request,” “HttpPost,” “HttpGet,” URL‐encode calls).

    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
     of
    -1.10
    kadang
    -1.05
     rozwiąz
    -0.95
     客厅
    -0.94
    interess
    -0.93
     Krzysz
    -0.93
    vider
    -0.93
     their
    -0.91
    nasium
    -0.89
    stant
    -0.89
    POSITIVE LOGITS
    himself
    1.06
    Epil
    1.05
     ViewHolder
    1.05
    ља
    1.03
    Exca
    1.02
     иначе
    1.01
    TestingModule
    1.00
    Liza
    0.97
    Requiem
    0.97
    icznego
    0.96
    Activations Density 0.015%

    No Known Activations