© 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. 12939
    Prev
    Next
    INDEX
    Explanations

    shape and dimension manipulation

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

    The neuron fires on mentions of “flatten” (i.e. code or function calls that flatten or un-nest data structures).

    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
     Such
    -1.84
    翃
    -1.83
    </i>
    -1.70
            
    -1.69
    衣服
    -1.64
    ꜆
    -1.64
     addirittura
    -1.63
     thậm
    -1.61
     содержит
    -1.60
     vermek
    -1.59
    POSITIVE LOGITS
    '
    2.16
    鱨
    1.89
    梣
    1.80
     yake
    1.74
    
    1.73
     ヌード
    1.71
    ,
    1.67
     素描
    1.62
     hins
    1.61
    ꔛ
    1.57
    Activations Density 0.001%

    No Known Activations