© 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. 1101
    Prev
    Next
    INDEX
    Explanations

    spectrum sunscreen

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

    This neuron detects technical specification terms—numbers paired with units or measurement-related adjectives in product or camera specs.

    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
    tpVar
    0.63
    Pension
    0.59
    Prosec
    0.59
    Democratic
    0.59
    INTa
    0.58
    Democrats
    0.57
     Обще
    0.56
    卍
    0.55
    Republican
    0.55
     行っ
    0.55
    POSITIVE LOGITS
     type
    0.65
     ("
    0.62
     mode
    0.58
     (
    0.57
     ('
    0.57
     (`
    0.57
     (=
    0.57
     $(
    0.57
     counterparts
    0.55
    /
    0.55
    Activations Density 0.632%

    No Known Activations