© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8067
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on uncommon content words—i.e. tokens that are semantically rich or low‐frequency in the text.

    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
    即使
    0.56
    LogError
    0.52
    Bre
    0.50
    Regardless
    0.48
    -[#
    0.48
    роят
    0.48
    }-\
    0.47
    adal
    0.47
     পক্ষে
    0.47
    ;\
    0.46
    POSITIVE LOGITS
     perhaps
    2.05
     eigenlijk
    1.77
    perhaps
    1.68
     basically
    1.60
     aslında
    1.55
     actually
    1.54
     eigentlich
    1.52
     quizás
    1.43
    其实
    1.42
     egent
    1.42
    Activations Density 0.296%

    No Known Activations