© 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. 5566
    Prev
    Next
    INDEX
    Explanations

    understanding categories

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

    The neuron strongly activates on uncommon, content-heavy words—especially technical or infrequent vocabulary.

    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
    아
    0.93
    اری
    0.84
    artney
    0.81
    어
    0.80
    ionali
    0.79
    라
    0.78
     hewan
    0.75
    َا
    0.75
    bolinks
    0.74
    anın
    0.74
    POSITIVE LOGITS
    dosage
    0.77
    更多的
    0.77
     incluyendo
    0.75
     applause
    0.75
    getInput
    0.74
     Spinning
    0.74
    Plugin
    0.73
     insist
    0.73
     বেশী
    0.73
     incluir
    0.73
    Activations Density 0.000%

    No Known Activations