© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7885
    Prev
    Next
    INDEX
    Explanations

    static fields

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

    This neuron detects Java-style fully qualified class names (dot-separated package and class identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    technical documentation or API reference text, particularly method definitions, file paths, and code documentation.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    technical programming documentation and stack-trace style text referencing classes, methods, exceptions, files, and “Definition at line … of file …” metadata

    oai_token-act-pair · gpt-5Triggered 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
     MyRegisterClass
    0.73
    btnLogout
    0.70
     Под
    0.69
     Quadrupèdes
    0.69
     под
    0.68
    SelectionSet
    0.68
    ၡ
    0.68
     विमान
    0.67
    班牙
    0.66
    oringa
    0.66
    POSITIVE LOGITS
     package
    0.71
     index
    0.65
     overview
    0.65
     apres
    0.61
     hierarchy
    0.61
     sell
    0.61
     packages
    0.61
     navigation
    0.60
     here
    0.59
     UI
    0.59
    Activations Density 0.005%

    No Known Activations