© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12442
    Prev
    Next
    INDEX
    Explanations

    tree structures and recursion

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

    This neuron activates on terms and contexts describing hierarchical or nested structures (e.g., hierarchy, subordinate/child pages, parent relationships).

    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
    вар
    0.52
    PasswordEncoder
    0.52
    间隔
    0.51
    駑
    0.51
     सरकारने
    0.50
    itania
    0.50
    醤
    0.50
     Лига
    0.49
    aqua
    0.48
     aru
    0.48
    POSITIVE LOGITS
     tree
    2.09
     subtree
    2.09
     Tree
    2.05
    subtree
    1.94
    Tree
    1.93
    TreeNode
    1.89
     TREE
    1.87
    tree
    1.84
     TreeNode
    1.84
    树
    1.80
    Activations Density 0.092%

    No Known Activations