© 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. 3945
    Prev
    Next
    INDEX
    Explanations

    words following "after"

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

    The neuron strongly activates on four-digit year mentions (especially when introduced by “After”).

    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
    ك
    1.85
    ع
    1.49
    ர்ஸ்
    1.47
    时候
    1.45
     itulah
    1.44
    ب
    1.40
    ف
    1.39
    пи
    1.38
    ী
    1.32
    بك
    1.31
    POSITIVE LOGITS
    words
    1.84
    taste
    1.80
    noons
    1.64
    s
    1.64
    glow
    1.63
     wards
    1.60
    effects
    1.59
    thoughts
    1.54
     পরই
    1.52
    effect
    1.35
    Activations Density 0.098%

    No Known Activations