© 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. 34-GEMMASCOPE-RES-131K
    7. 78582
    Prev
    Next
    INDEX
    Explanations

    up or up-regulated

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

    The neuron primarily detects occurrences of the token “up” (including variants like “Up”) that often appear in scientific terms (e.g. up-regulation, up-conversion).

    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
    が
    -0.81
    ですが
    -0.75
     major
    -0.73
    тнес
    -0.69
     seized
    -0.69
    宣言
    -0.68
     areas
    -0.67
    redirected
    -0.67
     at
    -0.66
    solr
    -0.66
    POSITIVE LOGITS
    upt
    1.06
    uppo
    0.99
    upto
    0.98
     Up
    0.96
     עד
    0.95
     DOWN
    0.95
    様な
    0.95
    を書く
    0.95
    Arrow
    0.93
     lên
    0.93
    Activations Density 0.024%

    No Known Activations