© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13600
    Prev
    Next
    INDEX
    Explanations

    This neuron is detecting rare or distinctive proper nouns and named entities (i.e. capitalized names and unique titles).

    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
    끗
    0.79
     две
    0.75
     gadget
    0.73
     dov
    0.72
    ევს
    0.72
     ಬೆ
    0.71
     bist
    0.71
    ટો
    0.71
     bluff
    0.70
     beginnings
    0.68
    POSITIVE LOGITS
    l
    0.89
    ι
    0.83
    h
    0.80
    r
    0.79
    ৫
    0.79
    €™
    0.79
    <s>
    0.78
    C
    0.77
    ξ
    0.77
    ق
    0.76
    Activations Density 0.000%

    No Known Activations