© 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. 5289
    Prev
    Next
    INDEX
    Explanations

    half of, rates and, modern songs, down/sideways, LE (lower extremity), stage

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

    The neuron is basically a “segment‐start” detector—it spikes on the very first content token of each new document snippet.

    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
    er
    0.92
    o
    0.80
    то
    0.75
    re
    0.74
    ta
    0.72
    et
    0.71
    to
    0.71
    FOL
    0.68
    ेशन
    0.68
    м
    0.68
    POSITIVE LOGITS
    ुरू
    0.89
     Tets
    0.83
     khủng
    0.79
     harmon
    0.79
     reacts
    0.77
     Narcotics
    0.74
     nark
    0.74
     khán
    0.73
     vírus
    0.73
    كتب
    0.73
    Activations Density 0.000%

    No Known Activations