© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 11147
    Prev
    Next
    INDEX
    Explanations

    formerly followed by an organization

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

    The neuron is not looking for semantic content so much as “special” tokens – it lights up on punctuation (exclamation points, apostrophes, quotes, colons), digit‐containing tokens (years, numerals, decimals), and capitalized/name fragments rather than ordinary lower-case words.

    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
    🫶
    1.38
    🩷
    1.22
    🥹
    1.16
    🫠
    1.11
    🤎
    1.09
    🫢
    1.09
    🥲
    1.07
    🤍
    1.05
     pandemic
    1.05
    🫧
    1.05
    POSITIVE LOGITS
     ২০১৮
    1.84
     ২০১৭
    1.72
    Brexit
    1.28
    Trump
    1.26
    🖒
    1.26
     ২০১৯
    1.24
     Brexit
    1.19
     Trump
    1.18
    特朗普
    1.00
     ২০১৬
    0.99
    Activations Density 0.891%

    No Known Activations