© 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. 49983
    Prev
    Next
    INDEX
    Explanations

    GNU General Public License version

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

    The neuron is highly responsive to tokens that are proper nouns or standalone numerals—i.e. any token beginning with an uppercase letter or consisting entirely of digits.

    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
    ApiController
    -0.79
     botto
    -0.78
    akuya
    -0.75
     dlc
    -0.74
     העולם
    -0.74
    ‌ای
    -0.73
     onyx
    -0.73
     pj
    -0.73
    ră
    -0.73
    setData
    -0.73
    POSITIVE LOGITS
     Metals
    0.80
    onto
    0.77
     磨
    0.75
    ór
    0.75
    Microphone
    0.74
    美子
    0.73
     Usted
    0.72
     カラー
    0.72
     zainteres
    0.72
    abetes
    0.71
    Activations Density 0.007%

    No Known Activations