© 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. 6906
    Prev
    Next
    INDEX
    Explanations

    The neuron is tuned to pick out chemical compound abbreviations (e.g. resin monomer names like “bis-GMA”).

    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
    Bonjour
    0.82
    With
    0.81
    Concent
    0.79
    Bij
    0.79
    SUN
    0.79
    While
    0.76
    Yeh
    0.75
    𝙔
    0.75
    Alabama
    0.74
    Dive
    0.74
    POSITIVE LOGITS
    ck
    0.83
    ack
    0.79
    чный
    0.79
     стрель
    0.78
    ленный
    0.78
    ii
    0.77
    iva
    0.77
    ik
    0.77
    ir
    0.76
    ica
    0.75
    Activations Density 0.000%

    No Known Activations