© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129966
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for high-frequency function words and punctuation (common stopwords like “but,” “to,” “never,” and sentence-ending periods).

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

    court clerk in legal documents

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    tedì
    -0.93
     cornu
    -0.89
    rha
    -0.88
     متوسط
    -0.86
    gató
    -0.85
    ？』
    -0.84
     carina
    -0.84
     jungles
    -0.83
     itself
    -0.83
     its
    -0.83
    POSITIVE LOGITS
     hum
    0.93
     (!__
    0.90
    úst
    0.87
    }%
    0.83
    deko
    0.83
     hason
    0.81
    Write
    0.81
     Serien
    0.80
    produced
    0.80
     piernas
    0.79
    Activations Density 0.068%

    No Known Activations