© 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. 12137
    Prev
    Next
    INDEX
    Explanations

    The neuron is strongly responsive to uncommon, content‐bearing words (i.e. tokens with high inverse-document-frequency) rather than common function words.

    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
    ៕
    0.69
     btw
    0.68
     again
    0.68
     ಮತ್ತೆ
    0.66
    ങ്കിലും
    0.66
     później
    0.64
    <start_of_image>
    0.63
    например
    0.63
     दोबारा
    0.62
    ?<
    0.62
    POSITIVE LOGITS
     Whilst
    0.89
    ↵
    0.86
     เนื่องจาก
    0.84
     Öncelikle
    0.83
     Notwithstanding
    0.82
     Sebagai
    0.74
    
    0.74
     Despite
    0.73
     這個
    0.73
     Несмотря
    0.73
    Activations Density 0.370%

    No Known Activations