© 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. 5909
    Prev
    Next
    INDEX
    Explanations

    The neuron activates strongly on proper nouns and numerical figures, essentially picking out specific names or numbers in the text.

    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
    ];//
    0.81
    Ꭻ
    0.81
    وج
    0.80
    فا
    0.79
    客户端
    0.78
    كتب
    0.77
    Collectors
    0.76
    充足
    0.76
    顧
    0.75
    );//
    0.73
    POSITIVE LOGITS
     hs
    0.85
     flare
    0.81
    িং
    0.78
    িক
    0.77
     mdl
    0.76
    tte
    0.75
     bilin
    0.75
    licer
    0.75
    ാ
    0.75
    শীল
    0.73
    Activations Density 0.000%

    No Known Activations