© 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. 22-GEMMASCOPE-RES-131K
    7. 25549
    Prev
    Next
    INDEX
    Explanations

    clinical, scientific, and unicode texts

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

    Unicode and character encoding references in technical text.

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

    This neuron responds to standalone numeric tokens (digits and hexadecimal numbers) 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
     get
    -1.49
     your
    -1.37
     started
    -1.32
     system
    -1.30
     time
    -1.29
    ִּ
    -1.27
    ּוֹ
    -1.23
    ַּ
    -1.18
     launched
    -1.15
     assessment
    -1.13
    POSITIVE LOGITS
     prêtre
    1.42
     malattie
    1.34
     religieux
    1.33
     religieuses
    1.30
     mentales
    1.27
     borbo
    1.25
     chré
    1.20
     ویتامین
    1.19
     árvores
    1.15
    了下去
    1.15
    Activations Density 0.003%

    No Known Activations