© 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. 2904
    Prev
    Next
    INDEX
    Explanations

    guidelines and guidance documents

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

    The neuron activates on words naming formal policy or instructional documents and resources (e.g. guidance, manuals, training modules, standards, legislation).

    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
     nasel
    -0.91
    ssysteme
    -0.89
     segun
    -0.80
    otras
    -0.78
    нительно
    -0.77
     haast
    -0.77
     populaires
    -0.76
    personalized
    -0.76
     Кубка
    -0.75
     žel
    -0.74
    POSITIVE LOGITS
     guidelines
    3.30
     guide
    3.16
     guidance
    2.92
     Guidelines
    2.48
     guides
    2.42
     guideline
    2.36
     Guide
    2.19
     toolkit
    2.16
     tool
    2.14
    Guidelines
    2.13
    Activations Density 0.069%

    No Known Activations