© 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. 31-GEMMASCOPE-2-RES-16K
    7. 15649
    Prev
    Next
    INDEX
    Explanations

    size and weight

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is triggered by mentions of specific bird species names (especially multiword, capitalized names like “Elegant Tern”).

    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
     invertebrate
    1.05
     invertebrates
    0.99
     hayvan
    0.96
     crustaceans
    0.95
     aquatic
    0.95
     vertebrate
    0.95
     hewan
    0.95
     aquaculture
    0.93
     vertebrates
    0.92
     वन्य
    0.92
    POSITIVE LOGITS
     elusive
    0.71
     ದಾಖ
    0.69
     bridesmaid
    0.65
     Congrats
    0.64
     impegno
    0.64
     finance
    0.64
     Finance
    0.64
     transaksi
    0.64
     Kigali
    0.63
     ند
    0.63
    Activations Density 0.217%

    No Known Activations