© 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. 11643
    Prev
    Next
    INDEX
    Explanations

    early April events

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

    The neuron fires on isolated, standalone words or short labels that serve as section headings or topic titles at the start of a new block.

    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
     Salaries
    0.75
    很好
    0.74
    不错
    0.74
    藩
    0.73
    不錯
    0.71
    妓
    0.71
    吊
    0.70
    遂
    0.69
     Ее
    0.68
    就不
    0.68
    POSITIVE LOGITS
     estes
    0.70
     elev
    0.69
    વ
    0.69
    log
    0.68
    stern
    0.67
     elast
    0.67
     अनुरूप
    0.66
    v
    0.66
     essa
    0.66
    tal
    0.65
    Activations Density 0.000%

    No Known Activations