© 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. 282
    Prev
    Next
    INDEX
    Explanations

    The neuron detects C‐style “for” loop headers (e.g. the tokens in `for (int i = …; i < …; ++i)`).

    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
     설립
    0.42
     联系
    0.41
    amélioration
    0.40
    奉
    0.40
     교육
    0.39
     作り
    0.39
    ATIVA
    0.39
     산업
    0.38
     Serviço
    0.38
    Ö
    0.38
    POSITIVE LOGITS
     x
    0.80
     idx
    0.80
     tmp
    0.79
     i
    0.75
     curr
    0.71
     _,
    0.69
     p
    0.68
     temp
    0.68
    idx
    0.68
     result
    0.66
    Activations Density 0.033%

    No Known Activations