© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12200
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to words describing garments that are snug or cling closely to the body, like “tight,” “fitting,” and “hugging.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     uncovered
    0.63
    解锁
    0.61
     ē
    0.57
     indirectly
    0.56
     swapped
    0.55
     liberated
    0.55
     unearthed
    0.55
     벗
    0.55
     breakaway
    0.55
     broken
    0.54
    POSITIVE LOGITS
    compact
    1.98
     compact
    1.84
     compactness
    1.83
    tight
    1.80
    Compact
    1.80
     compacted
    1.79
    Tight
    1.78
     compaction
    1.74
     компакт
    1.74
     tight
    1.73
    Activations Density 0.543%

    No Known Activations