© 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. 38350
    Prev
    Next
    INDEX
    Explanations

    comparisons or combinations

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

    The neuron strongly activates on specialized technical jargon—particularly scientific or biomedical terms (e.g. supernatant, memory [in a technical context], regions, stereospecific)—marking domain‐specific vocabulary.

    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
     berg
    -1.00
     bensin
    -0.93
    displayed
    -0.93
     kalender
    -0.92
     gunung
    -0.91
     asos
    -0.90
    yship
    -0.89
     leher
    -0.89
    rema
    -0.89
     kura
    -0.88
    POSITIVE LOGITS
     at
    1.02
     revital
    0.88
     consisting
    0.84
     scheduling
    0.83
     during
    0.82
     made
    0.81
    FXML
    0.80
    দ
    0.80
     until
    0.78
     становника
    0.78
    Activations Density 0.198%

    No Known Activations