© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6865
    Prev
    Next
    INDEX
    Explanations

    variety

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

    This neuron detects English words containing an apostrophe, such as contractions (e.g. doesn’t, what’s) and possessive ’s forms.

    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
     ಶೆ
    0.46
    ټبال
    0.46
     στον
    0.44
     lLoginID
    0.42
    sorption
    0.42
    𝓇
    0.42
     addCriterion
    0.40
    میم
    0.40
    𝘥
    0.39
    ϳ
    0.39
    POSITIVE LOGITS
     Covid
    0.47
     hefty
    0.46
     repurposed
    0.43
    Covid
    0.43
     debunk
    0.43
    ↵
    0.43
     COVID
    0.41
    ⠀
    0.41
    ↵↵
    0.40
    დე
    0.40
    Activations Density 0.002%

    No Known Activations