© 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. 98510
    Prev
    Next
    INDEX
    Explanations

    single content, unique elements

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

    This neuron highlights the main topic noun of each segment (e.g. “page,” “list,” “folder,” “profile,” “plot,” 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_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
     triage
    -0.83
     indiscrimin
    -0.80
     aggregate
    -0.79
     forego
    -0.76
     Ответить
    -0.76
    ウルトラ
    -0.75
    ัด
    -0.74
     μη
    -0.73
    索引
    -0.71
     劇場
    -0.71
    POSITIVE LOGITS
     singolo
    0.98
    單
    0.97
     riêng
    0.96
     solo
    0.96
     Einzel
    0.93
    SINGLE
    0.91
    单
    0.91
     länge
    0.90
    单个
    0.88
     отдельно
    0.88
    Activations Density 0.127%

    No Known Activations