© 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. 10-GEMMASCOPE-RES-131K
    4. 1783
    Prev
    Next
    INDEX
    Explanations

    code or article titles

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

    This neuron identifies “dunder” names—identifiers surrounded by double underscores (like __main__ or __esModule__).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    derdag
    -1.80
     Geräten
    -1.77
    搥
    -1.76
     namorados
    -1.71
    自身的
    -1.68
    汭
    -1.66
    嚶
    -1.63
     fazenda
    -1.59
    ]];
    -1.54
    éter
    -1.53
    POSITIVE LOGITS
    ↵
    1.93
    ↵↵
    1.86
    </i>
    1.76
    <bos>
    1.65
    </h5>
    1.63
    úgó
    1.61
    According
    1.55
     empêcher
    1.55
    Médaille
    1.53
    This
    1.52
    Activations Density 0.000%

    No Known Activations