© 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. 67044
    Prev
    Next
    INDEX
    Explanations

    years and anniversaries

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

    The neuron fires strongly on tokens written in full uppercase (e.g. headings, emphatic words, or “shouty” ALL-CAPS segments).

    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
     catalán
    -0.91
    遁
    -0.85
    aragua
    -0.85
    ánicos
    -0.84
    umbang
    -0.83
    Glej
    -0.82
     bekämp
    -0.82
     captive
    -0.82
     brasileiros
    -0.81
     ďal
    -0.81
    POSITIVE LOGITS
     its
    1.29
    Its
    0.93
     their
    0.89
     OUTER
    0.88
     Loco
    0.85
     ihrem
    0.83
     ``
    0.83
    gée
    0.82
     suoi
    0.81
     своем
    0.81
    Activations Density 0.176%

    No Known Activations