© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 881
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on rare or unusual subword pieces (especially punctuation-laden or out-of-vocabulary BPE tokens).

    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
    нцикло
    0.45
     SalesLT
    0.45
    ясплат
    0.43
    カイブ
    0.42
    ROff
    0.42
     Responder
    0.42
    ロッパ
    0.41
    𝘶
    0.41
    TaskPojo
    0.41
     চাহিল
    0.41
    POSITIVE LOGITS
     it
    0.62
     more
    0.51
     a
    0.50
     be
    0.50
     that
    0.50
    s
    0.50
     such
    0.49
     only
    0.49
     as
    0.48
     not
    0.47
    Activations Density 0.925%

    No Known Activations