© 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. 7712
    Prev
    Next
    INDEX
    Explanations

    numpy numerical operations

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

    The neuron fires on technical terms and parameters around computing an inverse FFT—e.g. “inverse,” “FFT,” array dimensions, axes, and related function‐call tokens.

    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
     franqu
    0.68
    岘
    0.65
     Falun
    0.64
     Vodafone
    0.63
     Gouver
    0.63
     Cadillac
    0.62
     сущ
    0.62
     Smack
    0.61
    ಕ್ಕ
    0.61
     envoyé
    0.60
    POSITIVE LOGITS
     numpy
    1.62
    numpy
    1.45
    [:,
    1.41
     np
    1.40
     dtype
    1.34
    ndarray
    1.34
     Numpy
    1.31
     scipy
    1.30
    dtype
    1.29
     NumPy
    1.27
    Activations Density 0.007%

    No Known Activations