© 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. 33025
    Prev
    Next
    INDEX
    Explanations

    writ denied or refused

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

    The neuron is most strongly activated by tokens that are part of document‐level metadata or section headings—especially those containing digits (e.g. version numbers, encoding labels) or single‐word headings like “Usage” or “Writ.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    document metadata and boilerplate headers/structure lines (e.g., usage statements, meta tags, legal citation/writ notations, dataset structure summaries).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    Looking at the activations, this neuron strongly activates for: - "Writ" (832, 832, 648) - Numbers like "82-" (236) in legal case citations - Words related to legal/formal document structure: "obs." (

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    FILTER
    -0.88
    펩
    -0.75
    Filter
    -0.73
    maps
    -0.73
     Wunder
    -0.73
    Remarks
    -0.72
     Contract
    -0.71
     FILTER
    -0.70
     comentarios
    -0.69
    tapan
    -0.68
    POSITIVE LOGITS
     ligation
    0.80
    ất
    0.69
    フラット
    0.69
    dealloc
    0.66
    ッソ
    0.66
    рации
    0.65
     startX
    0.65
     Roja
    0.65
     GTR
    0.64
     isotropic
    0.63
    Activations Density 0.076%

    No Known Activations