© 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. 34138
    Prev
    Next
    INDEX
    Explanations

    statement terminator

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

    The neuron primarily spots literal constants (numeric values, booleans like false, and null).

    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
     then
    -1.19
    onekana
    -1.05
     ngayon
    -1.02
     will
    -0.99
     only
    -0.98
     ansehen
    -0.98
    ですが
    -0.97
     from
    -0.96
     we
    -0.95
    からね
    -0.94
    POSITIVE LOGITS
    Poems
    1.00
    Gabriela
    0.95
    にい
    0.92
    異なる
    0.88
    +
    0.87
    Raleigh
    0.87
    ="_
    0.84
     عدم
    0.83
    dafri
    0.83
    ulet
    0.83
    Activations Density 0.004%

    No Known Activations