© 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. 13673
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly flags unusually rare or out‐of‐vocabulary tokens—i.e. words the model has almost never seen.

    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
    лү
    0.79
    ѵ
    0.73
    үз
    0.72
    ك
    0.70
    ocate
    0.69
    שׁ
    0.68
     وبعد
    0.66
    بعد
    0.66
     المقدس
    0.66
     গ
    0.65
    POSITIVE LOGITS
    t
    0.80
    m
    0.79
    hak
    0.76
    tour
    0.73
    แมนเชสเตอร์
    0.73
     Ego
    0.71
    juan
    0.70
    hams
    0.70
    talk
    0.70
    주는
    0.68
    Activations Density 0.000%

    No Known Activations