© 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. 10-GEMMASCOPE-RES-131K
    7. 20338
    Prev
    Next
    INDEX
    Explanations

    immediate commencement

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

    The neuron flags names of concrete entities or objects—especially technical/data‐structure terms or concept labels like “Dictionary,” “frames,” “tables,” “planes,” “product,” “Timeout,” etc.

    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
    参考にして
    -2.03
     then
    -1.80
     November
    -1.52
    𝗼
    -1.48
     😘
    -1.48
     December
    -1.41
     by
    -1.40
     подобрать
    -1.37
     چنین
    -1.35
    它们的
    -1.34
    POSITIVE LOGITS
     that
    1.69
     Majesté
    1.41
    ，
    1.41
    ì
    1.38
     probabilidad
    1.38
    cts
    1.34
     androidx
    1.34
    了
    1.34
     должны
    1.31
     сразу
    1.31
    Activations Density 0.192%

    No Known Activations