© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 107091
    Prev
    Next
    INDEX
    Explanations

    legal rights, audit documentation

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

    The neuron strongly activates on “audit” and other audit‐related terms (e.g. documentation, planning, work, evidence, review), i.e. it detects references to auditing processes and vocabulary.

    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
     Vecchio
    -0.75
    Stab
    -0.72
    andinavia
    -0.72
    kovic
    -0.71
     Stab
    -0.71
     basso
    -0.71
     Muhammad
    -0.70
    徕
    -0.70
    团体
    -0.70
    quiet
    -0.68
    POSITIVE LOGITS
     remuneration
    0.83
     verres
    0.80
    外し
    0.80
     Zou
    0.80
     प्रत्येक
    0.79
     sculptured
    0.79
     épaules
    0.79
    ሚ
    0.79
    υμ
    0.79
     ſaid
    0.79
    Activations Density 0.005%

    No Known Activations