© 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. 111118
    Prev
    Next
    INDEX
    Explanations

    Leapster, Maternal, Introduction, Li Lei, Active attacks

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

    The neuron strongly activates on tokens composed of digits or digit‐laden sequences (e.g. version numbers, years, or other numeric strings).

    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
    脍
    -0.84
     overcoat
    -0.80
    sidian
    -0.78
    执
    -0.75
     倒
    -0.71
     split
    -0.70
     Koper
    -0.69
    執
    -0.69
    êque
    -0.68
     SoC
    -0.68
    POSITIVE LOGITS
    blos
    0.89
    thou
    0.85
    graphe
    0.79
     страни
    0.79
     thou
    0.78
    × 
    0.77
     LOSS
    0.76
    bids
    0.74
     Leap
    0.73
    hydra
    0.73
    Activations Density 0.006%

    No Known Activations