© 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. 10-GEMMASCOPE-RES-131K
    7. 73816
    Prev
    Next
    INDEX
    Explanations

    electrical excitability

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

    technical or programming-related terminology, particularly related to JWT and configuration.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on isolated subword fragments (WordPiece tokens) that form parts of specialized technical or biomedical terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    姈
    -3.00
    并未
    -2.58
    骉
    -2.56
    
    -2.55
     всѣ
    -2.44
    secuencias
    -2.36
    producción
    -2.28
     ausschließlich
    -2.28
     masterful
    -2.27
    酲
    -2.27
    POSITIVE LOGITS
    塒
    3.05
    When
    2.78
    If
    2.70
    *
    2.58
    !'
    2.45
    ↓↓↓
    2.42
     стал
    2.41
    錞
    2.41
    to
    2.33
    www
    2.33
    Activations Density 0.000%

    No Known Activations