© 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. 10-GEMMASCOPE-RES-131K
    7. 127632
    Prev
    Next
    INDEX
    Explanations

    directional movement or position

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

    The neuron consistently activates on occurrences of the standalone word “down,” regardless of context.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    B
    -3.09
    J
    -2.81
    {
    -2.81
    桄
    -2.72
    」，
    -2.67
    >
    -2.58
     There
    -2.56
    G
    -2.48
    W
    -2.42
    S
    -2.41
    POSITIVE LOGITS
    偁
    3.14
    नलोड
    2.89
    妗
    2.84
     GOtt
    2.83
    ी
    2.80
    嫪
    2.75
    鷓
    2.75
     עם
    2.70
    嘞
    2.70
    ⢎
    2.66
    Activations Density 0.028%

    No Known Activations