© 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. 79017
    Prev
    Next
    INDEX
    Explanations

    one in nine, 6% of

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

    The neuron activates on numerical tokens, especially multi‐digit figures and statistics.

    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
    orghe
    -0.75
    linha
    -0.75
    =$_
    -0.73
     三
    -0.72
     done
    -0.72
    我还是
    -0.71
     mož
    -0.71
    lox
    -0.70
    remont
    -0.70
     Có
    -0.70
    POSITIVE LOGITS
    ¹,
    0.80
    ずに
    0.78
    倞
    0.76
    ,¹
    0.75
     gagne
    0.75
     garantit
    0.73
     Wert
    0.73
    ↹
    0.72
    
    0.72
    ाद
    0.72
    Activations Density 0.065%

    No Known Activations