© 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. 13677
    Prev
    Next
    INDEX
    Explanations

    This neuron spikes only on unusually rare or content-heavy words—i.e. tokens that occur infrequently and carry strong “topic” or semantic weight.

    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
    offe
    0.76
     khử
    0.74
    Ộ
    0.71
     கொடுத்து
    0.71
     accompagner
    0.68
     চরিত্রে
    0.67
    och
    0.66
    ading
    0.66
    }).
    0.65
     お知らせ
    0.65
    POSITIVE LOGITS
    Effect
    0.84
    myItem
    0.84
    کن
    0.82
    ступает
    0.77
    DidLoad
    0.77
     ligera
    0.75
     empezó
    0.75
    стал
    0.75
    ธาน
    0.74
    素晴
    0.73
    Activations Density 0.000%

    No Known Activations