© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71167
    Prev
    Next
    INDEX
    Explanations

    shape and size

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

    The neuron fires strongly on physical‐attribute terms—especially the word “shape” (and similar dimensional descriptors).

    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
     $^{\
    -1.84
    neling
    -1.71
    vení
    -1.58
    鱿
    -1.56
    を表示
    -1.55
    がこの
    -1.55
    ことです
    -1.53
    }^{\
    -1.52
     Not
    -1.52
     "
    -1.52
    POSITIVE LOGITS
     shapes
    1.91
     shape
    1.84
     minimalista
    1.78
    klere
    1.77
    หรือ
    1.72
     djur
    1.70
     öns
    1.68
     pessoas
    1.67
     fön
    1.66
    躄
    1.66
    Activations Density 0.059%

    No Known Activations