© 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. 11143
    Prev
    Next
    INDEX
    Explanations

    AECOPD

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

    The neuron is looking for non‐function “content” words—i.e. rarer, information‐bearing tokens rather than common stop words.

    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
    tun
    0.75
     ])
    0.73
    cant
    0.72
    fires
    0.71
     রকম
    0.70
     सिप
    0.70
    DEPEND
    0.70
    သို့
    0.70
    took
    0.70
    deploy
    0.69
    POSITIVE LOGITS
    یع
    0.82
    玮
    0.78
    संपादन
    0.75
     Bootcamp
    0.75
     pyperclip
    0.75
    ی
    0.74
     বয়
    0.74
     supplementation
    0.73
    мна
    0.73
     journaling
    0.73
    Activations Density 0.000%

    No Known Activations