© 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. 24-GEMMASCOPE-2-RES-16K
    7. 5752
    Prev
    Next
    INDEX
    Explanations

    `(` followed by `the`

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

    The neuron strongly activates on domain‐specific technical terms and multiword jargon (e.g. “hypodescent rule,” “gap analysis,” “Dynamic Lighting”).

    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_24_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
    했으며
    0.80
     dejó
    0.78
     resultó
    0.75
     secretário
    0.75
     beserta
    0.74
     ditambah
    0.73
     falleció
    0.72
    しましたが
    0.70
     tomó
    0.69
     했습니다
    0.69
    POSITIVE LOGITS
     یعنی
    1.20
    つまり
    1.15
     which
    1.13
     whereby
    1.13
    یعنی
    1.10
    也就是说
    1.09
    which
    1.08
     yani
    1.03
     অর্থাৎ
    1.02
     которое
    1.01
    Activations Density 0.327%

    No Known Activations