© 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. 25417
    Prev
    Next
    INDEX
    Explanations

    arithmetic operations and assignments

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

    The neuron never activates on any token—it’s essentially “dead” and doesn’t detect any pattern.

    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
    times
    -1.02
    resulting
    -0.91
    almost
    -0.88
    嗑
    -0.88
     ABBREVIATIONS
    -0.88
    nearly
    -0.87
    possibly
    -0.86
    ranno
    -0.85
    urface
    -0.85
    irmanship
    -0.84
    POSITIVE LOGITS
     encantó
    0.91
     one
    0.87
     Liege
    0.87
    💫
    0.85
     around
    0.85
    mẫu
    0.85
     harian
    0.84
    OCOL
    0.83
     rtn
    0.82
    可以
    0.82
    Activations Density 0.001%

    No Known Activations