© 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. 13257
    Prev
    Next
    INDEX
    Explanations

    code or script syntax

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

    The neuron detects apostrophe-led contraction fragments (e.g. ’re, ’ve, ’s) in text.

    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
     Cf
    0.57
    Cf
    0.53
     なく
    0.53
     corroborate
    0.52
    ̉
    0.48
     Excluding
    0.48
     cause
    0.47
     proposition
    0.46
     bet
    0.46
     格
    0.46
    POSITIVE LOGITS
     '${
    0.57
     акча
    0.57
     pauper
    0.56
     encamin
    0.55
     lãi
    0.54
    bereit
    0.54
    кому
    0.53
    եցին
    0.53
     रक्कम
    0.52
     plush
    0.52
    Activations Density 0.029%

    No Known Activations