© 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. 7198
    Prev
    Next
    INDEX
    Explanations

    boundary delimiters or code separation

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

    The neuron is a “rare‐token” detector—it spikes on unusually infrequent or low‐frequency subword tokens (proper nouns, URLs, uncommon words).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    email addresses and web URLs.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    indicators of web or online content and site metadata, such as links, email addresses, and blog/page structural cues.

    oai_token-act-pair · gpt-5Triggered 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
     connus
    0.94
    状况
    0.88
     américains
    0.84
    策略
    0.82
    য়িক
    0.81
    缱
    0.80
    现有
    0.80
    ра
    0.80
    botones
    0.79
    情況
    0.79
    POSITIVE LOGITS
    ‟
    0.72
     Prepar
    0.71
     alumna
    0.70
     আর্ম
    0.69
     extrema
    0.68
     Papua
    0.68
     arugula
    0.66
     Lim
    0.66
     autopsy
    0.66
     Cit
    0.65
    Activations Density 0.005%

    No Known Activations