© 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. 8764
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on tokens that are names or identifiers—especially acronyms, product‐ or model‐style strings, and other proper‐noun‐like labels.

    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
    ↵↵
    0.86
    1
    0.85
    4
    0.81
              
    0.78
     plight
    0.77
    0
    0.75
    2
    0.75
     Arrest
    0.74
    6
    0.73
    },
    0.71
    POSITIVE LOGITS
     necessitates
    0.80
     потребу
    0.78
    <unused23>
    0.76
     방식
    0.75
     transacción
    0.74
     pemrograman
    0.73
     reimb
    0.72
     prototyping
    0.72
     의해
    0.71
     смысле
    0.71
    Activations Density 0.001%

    No Known Activations