© 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. 42323
    Prev
    Next
    INDEX
    Explanations

    bland, boring, drab

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

    The neuron fires strongly on the adjective “bland” (and to a lesser extent similar dull descriptors), i.e. it detects calls of something being uninteresting or flavorless.

    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
     แต่
    -3.73
     ส์
    -3.27
    
    -3.05
    푢
    -2.97
    噥
    -2.83
     ſever
    -2.77
    因为
    -2.69
    蹰
    -2.67
     ་
    -2.56
     みた
    -2.56
    POSITIVE LOGITS
    “
    3.50
    O
    3.42
    which
    3.14
    .
    3.05
    D
    2.97
    E
    2.94
    2
    2.83
    /
    2.81
    X
    2.73
    1
    2.69
    Activations Density 0.001%

    No Known Activations