© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 75250
    Prev
    Next
    INDEX
    Explanations

    additional information or requirements

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

    The neuron fires on words used as labels or headings that introduce supplementary or “extra” information (e.g. “Additional,” “note,” “args,” “safeguards,” “considerations,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     robar
    -1.12
     имел
    -1.03
    🍗
    -0.95
     renouvel
    -0.95
     Hinsicht
    -0.94
     iconTwitter
    -0.93
    🍦
    -0.93
    相片
    -0.92
    Tenure
    -0.91
     ovation
    -0.91
    POSITIVE LOGITS
     than
    1.66
     added
    1.23
     дополнительные
    1.20
     дополнительных
    1.16
     in
    1.12
    beyond
    1.09
    extra
    1.05
     extra
    1.05
     zusätzliche
    1.04
     pueden
    1.02
    Activations Density 0.039%

    No Known Activations