© 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. 16338
    Prev
    Next
    INDEX
    Explanations

    n=1 assignment

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

    The neuron fires on semantically rich “content” tokens (nouns, verbs, adjectives, etc.) rather than on common function 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
    Chromatic
    0.88
    ']:
    0.80
    Lua
    0.80
    ফুদ্দিন
    0.79
    мыш
    0.79
    ധികം
    0.78
    gles
    0.77
    dummies
    0.75
     следующих
    0.75
    ʦ
    0.74
    POSITIVE LOGITS
    あれ
    0.83
    ク
    0.79
    度
    0.78
    管
    0.77
     Ev
    0.76
    かなり
    0.75
    歓
    0.75
    ア
    0.75
     Rembrandt
    0.73
    き
    0.73
    Activations Density 0.000%

    No Known Activations