© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111506
    Prev
    Next
    INDEX
    Explanations

    movement over ground or through paths

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

    The neuron responds to word stems that signal reversal or repetition—e.g. parts of “retrace/ retraced,” “retouch,” “back,” “criss-cross,” etc.

    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
    .
    -2.03
    ↵↵
    -1.94
    "
    -1.77
    \
    -1.76
    {
    -1.71
     is
    -1.69
     of
    -1.61
    -
    -1.61
    <b>
    -1.59
    ()
    -1.56
    POSITIVE LOGITS
     leone
    1.97
     mentre
    1.80
     perfons
    1.74
     水彩
    1.67
     possibili
    1.66
    𝗲
    1.66
     necessari
    1.63
    
    1.63
     土産
    1.62
     cristian
    1.61
    Activations Density 0.014%

    No Known Activations