© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 9447
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and code syntax

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

    The neuron fires on uppercase acronymic or trademark‐style tokens (e.g. AVAILABILITY, Layers, trademark, FC, SC).

    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
    ()
    1.20
    {
    1.10
    }
    1.05
    }{\
    0.99
    _{
    0.98
    ){
    0.96
    }\,
    0.96
    )
    0.94
    (){
    0.94
    {}
    0.94
    POSITIVE LOGITS
     récit
    0.92
     maíz
    0.87
     diversión
    0.86
     vendeurs
    0.86
     memnun
    0.86
     sărb
    0.84
     reuniones
    0.83
     república
    0.83
     entretenimiento
    0.82
     henkil
    0.82
    Activations Density 0.001%

    No Known Activations