© 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. 54532
    Prev
    Next
    INDEX
    Explanations

    android xml code

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

    This neuron activates on programming “method call” tokens (e.g. method names like with, load, from, into) in code.

    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
     seny
    -1.26
     régime
    -1.17
    ここから
    -1.16
     perda
    -1.12
     onPostExecute
    -1.09
     encontrou
    -1.08
     violon
    -1.07
     実例
    -1.07
    Rebs
    -1.06
     intensif
    -1.05
    POSITIVE LOGITS
    cenza
    1.20
     They
    1.16
     техні
    0.98
     betale
    0.98
    xxi
    0.97
     розта
    0.96
    rosted
    0.94
     OF
    0.94
    大丈夫です
    0.94
     навіть
    0.93
    Activations Density 0.002%

    No Known Activations