© 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. 70876
    Prev
    Next
    INDEX
    Explanations

    programming interfaces and patterns

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

    The neuron is detecting references to programming types and APIs (e.g. class, interface, event, observer, listener, builder, timer task, view) 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
     необхід
    -0.99
    гам
    -0.96
     Gunakan
    -0.94
     maravilla
    -0.92
    Vehiculo
    -0.90
     зовні
    -0.90
     Badezimmer
    -0.89
     如果
    -0.88
     Milliarden
    -0.88
     зберіга
    -0.88
    POSITIVE LOGITS
    UnifiedTopology
    0.92
     jupe
    0.90
    вости
    0.90
    Discus
    0.89
    フォー
    0.88
    Eb
    0.86
     keluarganya
    0.85
     genauso
    0.83
     télécommande
    0.82
     famille
    0.81
    Activations Density 0.001%

    No Known Activations