© 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. 13195
    Prev
    Next
    INDEX
    Explanations

    thus

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

    This neuron detects the technical phrasing of “aligning” or “overlapping” layer patterns (e.g. upper/lower overlay patterns) in patent‐style method descriptions.

    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
     meditations
    0.81
     arquivo
    0.78
     بكل
    0.73
     Nhưng
    0.72
    ちゃんと
    0.72
     anum
    0.71
     abbia
    0.71
     אבל
    0.71
     作詞
    0.70
     pero
    0.69
    POSITIVE LOGITS
     thereby
    1.33
    从而
    1.20
    导致
    1.07
    これにより
    1.05
    进而
    1.02
    会导致
    0.96
     Thereby
    0.96
     باعث
    0.91
     способствует
    0.89
     وبالتالي
    0.88
    Activations Density 3.404%

    No Known Activations