© 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. 99713
    Prev
    Next
    INDEX
    Explanations

    denied, canceled, affirmed, granted

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

    The neuron activates on formal decision‐verdict words in legal rulings (e.g. denied, affirmed, ordered, canceled).

    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
    ㍑
    -1.61
     flexibel
    -1.45
    鲠
    -1.36
     Cette
    -1.34
     essencial
    -1.34
    釔
    -1.33
     treff
    -1.33
     essentiel
    -1.32
     remarquable
    -1.31
    回到了
    -1.30
    POSITIVE LOGITS
     first
    2.23
     there
    1.74
     "
    1.70
     most
    1.62
     until
    1.58
     without
    1.56
     although
    1.53
     team
    1.33
    "
    1.32
     name
    1.27
    Activations Density 0.042%

    No Known Activations