© 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. 130427
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on tokens that are capitalized or numeric—i.e. proper nouns, acronyms, and standalone numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    listings code

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     however
    -1.39
     jednak
    -1.24
     echter
    -1.10
    however
    -1.05
     však
    -1.02
    stående
    -1.02
     and
    -1.00
     KeyCode
    -1.00
    ibatis
    -0.97
     هذه
    -0.97
    POSITIVE LOGITS
     thereof
    1.19
     are
    1.18
     его
    1.14
     celui
    1.10
     contactez
    1.05
     ceux
    1.03
     it
    1.00
     ones
    0.99
     were
    0.98
    )
    0.98
    Activations Density 0.021%

    No Known Activations