© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 65755
    Prev
    Next
    INDEX
    Explanations

    outlander

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

    The neuron strongly activates on tokens that are entirely uppercase (including acronyms, all‐caps words, and standalone numbers) as found in headings or titles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ogrod
    -0.73
    Здравствуйте
    -0.73
    refundable
    -0.73
    ZP
    -0.73
     recip
    -0.73
     Kok
    -0.72
    Gloria
    -0.72
    AVAN
    -0.71
    rowse
    -0.71
     Gloria
    -0.70
    POSITIVE LOGITS
     outlander
    1.13
    lander
    1.13
     Drought
    1.09
     Fraser
    1.05
     Out
    1.05
     Claire
    0.98
    Out
    0.92
    Claire
    0.90
     Jamie
    0.82
     MacKenzie
    0.82
    Activations Density 0.021%

    No Known Activations