© 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. 12687
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on words that kick off explanatory or definitional clauses—terms like “explains,” “incorporated,” “object,” or “From” that introduce a definition or description.

    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
    2
    0.59
    two
    0.57
    ↵
    0.56
    0
    0.54
    和
    0.54
    1
    0.53
    로
    0.52
     ਅਤੇ
    0.52
     illetve
    0.52
     .
    0.51
    POSITIVE LOGITS
     inexplicable
    0.58
     embank
    0.57
     oddly
    0.56
     ethos
    0.56
     unprofitable
    0.56
     mishaps
    0.55
     supremac
    0.55
     perverse
    0.55
     inept
    0.54
     clandestine
    0.54
    Activations Density 0.000%

    No Known Activations