© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11390
    Prev
    Next
    INDEX
    Explanations

    daily life

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

    The neuron activates on transliterated Greek lemma or inflectional morpheme tokens (e.g. ē, ōn, aiōn) indicating it’s detecting Greek word fragments or endings.

    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
     powerAll
    0.69
     zuk
    0.68
     abhis
    0.67
     Handsome
    0.66
    <unused58>
    0.64
     💞
    0.64
     baseHP
    0.64
     Jeśli
    0.63
     kabhi
    0.63
     కృ
    0.62
    POSITIVE LOGITS
    ́
    0.68
    ι
    0.63
    2
    0.55
    3
    0.54
    ing
    0.53
    6
    0.52
     due
    0.51
    going
    0.50
    _);
    0.48
    ν
    0.48
    Activations Density 0.019%

    No Known Activations