© 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. 8741
    Prev
    Next
    INDEX
    Explanations

    formal logic and systems

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

    The neuron fires on formal proof‐style or mathematical specification language—phrases stating definitions, theorems, numbered constraints, and proof validity assertions.

    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
     streamwise
    0.72
    convolution
    0.69
     agrupación
    0.66
     azimuthal
    0.63
    散热
    0.59
     yakın
    0.59
    样品
    0.58
    盆
    0.58
     convection
    0.56
     রাজশাহী
    0.56
    POSITIVE LOGITS
     semantics
    0.95
     predicates
    0.94
    vdash
    0.93
     propositional
    0.93
     predicate
    0.92
     syntactic
    0.90
    Predicate
    0.87
     axioms
    0.87
     विधे
    0.85
    predicate
    0.84
    Activations Density 0.116%

    No Known Activations