© 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. 14168
    Prev
    Next
    INDEX
    Explanations

    sentence ending punctuation

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

    This neuron detects the common discourse markers that introduce key factual statements or definitions in technical or academic prose (for example, “It is…,” “This study shows that…,” 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
     ـ
    -1.22
     ilustracji
    -1.09
    GRAFÍA
    -0.93
     Cette
    -0.92
     Malgré
    -0.92
    GRAFIA
    -0.91
     มิ
    -0.91
     Depuis
    -0.91
     dupla
    -0.90
     شركة
    -0.90
    POSITIVE LOGITS
     says
    1.21
     хочет
    1.09
     onSubmit
    1.02
     furthermore
    1.01
     pueden
    0.99
     cimetière
    0.98
    PYX
    0.97
    agreb
    0.97
    enosine
    0.94
     there
    0.94
    Activations Density 0.062%

    No Known Activations