© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 35748
    Prev
    Next
    INDEX
    Explanations

    thread witch

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

    The neuron activates on technical identifiers and code‐related names (e.g. tag names, class names, module names) in programming 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
    如果不是
    -0.81
    niająca
    -0.80
     Interesse
    -0.80
     aspirations
    -0.79
     dimentic
    -0.78
     Vereinigte
    -0.78
    寺の
    -0.77
    nemies
    -0.77
    september
    -0.74
    destinations
    -0.74
    POSITIVE LOGITS
     accepts
    0.91
     attends
    0.89
     eingeführt
    0.85
     restante
    0.82
     Figs
    0.82
    ility
    0.81
    口碑
    0.79
     břez
    0.79
    新冠
    0.79
     replaces
    0.79
    Activations Density 0.080%

    No Known Activations