© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 131019
    Prev
    Next
    INDEX
    Explanations

    despite followed by contrast

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

    The neuron fires strongly on the word “despite,” i.e. concessive constructions indicating contrast.

    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
    на
    -2.33
     also
    -2.16
     *
    -2.08
    ↵↵
    -2.05
    ↵↵↵
    -1.94
     of
    -1.91
    </sup>
    -1.88
    _{
    -1.88
    )$.
    -1.86
     '
    -1.85
    POSITIVE LOGITS
    襯
    2.17
     avancer
    2.02
     cuadros
    2.02
    生產
    1.98
     było
    1.95
    錞
    1.95
     visibilidad
    1.95
     zoude
    1.94
    一邊
    1.91
     sicheren
    1.91
    Activations Density 0.008%

    No Known Activations