© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5658
    Prev
    Next
    INDEX
    Explanations

    do/does followed by the/not/my

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

    The neuron strongly activates on emphatic uses of the auxiliary “do,” i.e. “do” used to add emphasis in a sentence.

    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
    ెస్
    1.16
    ᄊ
    1.13
     diciamo
    1.04
    CODES
    1.02
    İN
    1.02
    드시
    1.01
    ound
    1.00
    Edition
    1.00
     gouttes
    0.99
     çal
    0.99
    POSITIVE LOGITS
    ל
    1.43
     wonders
    1.33
     needful
    1.23
    ppel
    1.22
    omsday
    1.19
    lukan
    1.13
    zens
    1.12
    hány
    1.11
    鹃
    1.11
    er
    1.09
    Activations Density 0.303%

    No Known Activations