© 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. 8391
    Prev
    Next
    INDEX
    Explanations

    `ve` followed by past participles or `provided`

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

    This neuron is keyed to English clitic contractions (e.g. ’d, ’ve, ’m, n’t, it’s).

    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
    y
    1.70
    u
    1.57
    ek
    1.53
    oo
    1.48
    wv
    1.48
    wj
    1.48
    w
    1.45
    yj
    1.41
    yer
    1.38
    mht
    1.38
    POSITIVE LOGITS
    т
    1.66
    ТЕ
    1.36
    ım
    1.20
     gotta
    1.19
    ियल
    1.16
     Każ
    1.08
    за
    1.06
    ArgumentParser
    1.06
    ParentElement
    1.06
    িয়
    1.05
    Activations Density 3.966%

    No Known Activations