© 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. 34-GEMMASCOPE-RES-131K
    4. 2181
    Prev
    Next
    INDEX
    Explanations

    nuclear fuel disposal

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

    The neuron is a boilerplate‐detector that fires strongly on common function words and keywords found in legal/license and other repetitive “housekeeping” sections (e.g. Terms of Use, copyright/license notices, import headers), effectively flagging blocks of boilerplate 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_34/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
     sounds
    -1.05
     found
    -0.82
     Sounds
    -0.78
    这么说
    -0.76
    ('/:
    -0.75
    uala
    -0.73
    餾
    -0.72
     suena
    -0.72
    Sounds
    -0.71
    found
    -0.70
    POSITIVE LOGITS
    Ernst
    0.85
    ▹
    0.79
     guan
    0.77
    tilli
    0.76
     herbicide
    0.76
    あれば
    0.75
    giques
    0.74
    Preheat
    0.74
    Grill
    0.73
     荒
    0.73
    Activations Density 0.012%

    No Known Activations