© 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. 34-GEMMASCOPE-RES-131K
    7. 76415
    Prev
    Next
    INDEX
    Explanations

    legal contexts, exempli gratia

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

    The neuron activates on legal‐style provisions that assign burdens, obligations, or priority rights (e.g. “imposed burden on the accused to prove…” or rules about mortgage priority).

    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
     officiellement
    -1.02
     Moslem
    -0.98
     emir
    -0.94
    Confirmation
    -0.91
     tył
    -0.89
    îner
    -0.88
     serif
    -0.88
    付出
    -0.88
     Loa
    -0.88
     offiziell
    -0.88
    POSITIVE LOGITS
     remedy
    1.33
    liwości
    1.02
    辄
    1.02
    )=>{
    1.00
    joje
    0.99
    ρθ
    0.98
    {
    0.97
     prima
    0.96
    tlere
    0.96
    éte
    0.95
    Activations Density 0.019%

    No Known Activations