© 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. 5058
    Prev
    Next
    INDEX
    Explanations

    going to / gonna / will

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on first‐person plural references—especially “we” and “we’re” constructions signaling collective statements.

    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
     সমিত
    0.56
     olabilir
    0.56
     হওয়া
    0.54
     ਕੋ
    0.54
     indium
    0.52
     bi
    0.52
    髋
    0.51
     olursa
    0.49
     मौजूद
    0.49
    铍
    0.49
    POSITIVE LOGITS
     gonna
    1.79
     Gonna
    1.59
     going
    1.27
    going
    1.15
     Going
    1.05
    Going
    1.00
    GOING
    0.91
     goin
    0.89
     যাচ্ছে
    0.87
     जाणार
    0.86
    Activations Density 0.273%

    No Known Activations