© 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. 443
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on capitalized named entities—especially geographic place names (cities/regions) and similarly styled product or brand names.

    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
    
    0.44
     अद्भुत
    0.43
     lucrative
    0.43
    ekwondo
    0.42
     말미암
    0.41
     heartache
    0.41
     astounding
    0.41
     wykorzyst
    0.40
     heartwarming
    0.40
    rdoba
    0.40
    POSITIVE LOGITS
    ↵
    0.44
    is
    0.40
     etc
    0.38
     \
    0.37
    ie
    0.36
    m
    0.35
    adim
    0.33
    v
    0.33
    V
    0.33
     Se
    0.33
    Activations Density 0.221%

    No Known Activations