© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5204
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the pronoun “it,” especially when “it” begins a clause (e.g. “it is,” “it was,” “it called,” etc.).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     کدام
    0.71
     ఇతర
    0.70
     (=
    0.68
     (!)
    0.67
     आधारित
    0.67
     célèbre
    0.64
    (!)
    0.63
     některé
    0.62
    ฯ
    0.61
     আরেক
    0.60
    POSITIVE LOGITS
     simply
    0.88
     variously
    0.87
     colloqu
    0.81
     lihtsalt
    0.78
     herein
    0.78
     jednostavno
    0.75
     affectionately
    0.75
     просто
    0.74
     by
    0.73
     communément
    0.69
    Activations Density 0.228%

    No Known Activations