© 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. 3512
    Prev
    Next
    INDEX
    Explanations

    quality and heritage

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

    The neuron fires strongly on commercial or promotional business‐listing language—e.g. mentions of companies, “based,” “offering,” “services,” “established,” 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
     parsed
    0.70
    写法
    0.69
     получится
    0.67
     lesz
    0.66
     парла
    0.65
     ingested
    0.62
    ⁸
    0.62
    program
    0.61
     ప్రభావ
    0.61
    ⁴
    0.61
    POSITIVE LOGITS
     Founded
    1.12
    創業
    1.12
    經營
    1.10
     Manufacturer
    1.09
     품질
    1.09
     headquartered
    1.09
    成立于
    1.05
     Excellence
    1.05
    是一家
    1.05
     manufacturer
    1.03
    Activations Density 1.237%

    No Known Activations