© 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. 35010
    Prev
    Next
    INDEX
    Explanations

    medical treatment and research

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

    text related to scientific or technical documentation, especially when referring to measurements, patients, or systems.

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

    The neuron is primarily detecting numeric tokens (cardinal numbers and measurements) in the text.

    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
     която
    -0.99
    ５
    -0.98
    又是一
    -0.91
     kulinar
    -0.87
     نادي
    -0.87
    ไม่
    -0.85
    ３
    -0.83
    ᥴ
    -0.82
    別の
    -0.82
     Khusus
    -0.81
    POSITIVE LOGITS
     aimer
    0.89
     fourn
    0.88
     a
    0.88
     chapéu
    0.87
     essayer
    0.87
    quently
    0.81
     contrô
    0.80
    pineapple
    0.79
    icona
    0.79
    copies
    0.77
    Activations Density 0.407%

    No Known Activations