© 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. 31-GEMMASCOPE-2-RES-16K
    7. 984
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on capitalized proper nouns (names of people, organizations, places, events, brands, etc.).

    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_31_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
    CartVO
    0.58
    ঽ
    0.55
    TimeDict
    0.53
     ግራም
    0.53
    jumlah
    0.53
    MongoClient
    0.52
    VerFile
    0.52
    EventManager
    0.52
    marshalN
    0.52
    italizationType
    0.50
    POSITIVE LOGITS
    ®
    0.93
    ™
    0.90
     shenanigans
    0.80
    ®.
    0.79
     Syndrome
    0.79
    ®,
    0.77
     bandwagon
    0.74
     sightings
    0.73
     című
    0.72
    ™.
    0.72
    Activations Density 1.396%

    No Known Activations