© 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. 7096
    Prev
    Next
    INDEX
    Explanations

    framework, castle, Pull Request, version, Llama

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

    The neuron fires most strongly on unusual or out-of-vocabulary tokens—particularly abbreviations and rare named/entity-like strings.

    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
     Divid
    0.94
    ेक्स
    0.94
     Administ
    0.90
    ＮＡ
    0.88
    डी
    0.81
    Nama
    0.80
    て
    0.79
    イ
    0.78
    我們會
    0.77
    Div
    0.76
    POSITIVE LOGITS
    ামুটি
    0.79
    polling
    0.77
    rencies
    0.77
    ژگی
    0.77
    me
    0.74
     weir
    0.73
     fabs
    0.72
    ommen
    0.72
    чкой
    0.71
    ötz
    0.71
    Activations Density 0.001%

    No Known Activations