© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9262
    Prev
    Next
    INDEX
    Explanations

    Choose the best option

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

    This neuron strongly fires on pronouns or the definite article when they appear clause-initially (e.g. “they,” “we,” “the,” “it” right after a sentence break or comma).

    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
     ،
    0.47
     hoá
    0.46
    ᆢ
    0.46
     যাবত
    0.45
     照片
    0.45
    。。。。
    0.45
     nalazi
    0.44
     hrs
    0.44
     piac
    0.43
     ঃ
    0.42
    POSITIVE LOGITS
    ពួកគេ
    0.48
    但他
    0.44
     вони
    0.43
    ాతీయ
    0.43
    ціона
    0.43
     nonetheless
    0.42
    They
    0.41
     restre
    0.41
     they
    0.41
    Nonetheless
    0.40
    Activations Density 0.091%

    No Known Activations