© 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. 11299
    Prev
    Next
    INDEX
    Explanations

    ending or beginning phrases

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

    The neuron fires almost exclusively on those sporadic, abnormally large activation spikes—i.e. on out-of-vocabulary or otherwise unusual tokens that have been assigned anomalously high activation values.

    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
    Dispatcher
    0.75
     टक
    0.74
     nonchal
    0.74
    祓
    0.73
     Brawl
    0.72
    <0xD2>
    0.71
     depots
    0.71
     Batteries
    0.70
     inept
    0.70
    HintPath
    0.69
    POSITIVE LOGITS
    ת
    0.78
    ר
    0.77
     panjang
    0.75
    ي
    0.75
     Nachdem
    0.74
     puluh
    0.74
     soal
    0.73
    だ
    0.72
     spazi
    0.70
    ल्यानंतर
    0.69
    Activations Density 0.000%

    No Known Activations