© 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. 34191
    Prev
    Next
    INDEX
    Explanations

    Orangutan

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

    scientific or technical text with specialized terminology.

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

    The neuron activates on decimal numbers (numeric tokens containing a period).

    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
     química
    -1.11
     twenty
    -1.10
     rękaw
    -1.09
     против
    -1.07
     twelve
    -1.07
     two
    -1.07
     treating
    -1.05
     químico
    -1.03
     ใหญ่
    -1.02
     beż
    -1.02
    POSITIVE LOGITS
     型
    1.10
    ῖ
    1.10
    ô
    1.10
    ан
    1.07
    lí
    1.05
    ठ
    1.02
    su
    1.01
     обу
    1.00
    经常
    0.99
     pernic
    0.99
    Activations Density 0.000%

    No Known Activations