© 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. 9387
    Prev
    Next
    INDEX
    Explanations

    of crystals

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

    The neuron strongly activates on domain‐specific scientific or technical terminology—especially chemical formulas and compound names, mineral names, crystal‐structure headings, and related lab/paper section labels.

    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
    禿
    0.72
     immunotherapy
    0.64
     Luke
    0.61
     आर्क
    0.60
    Luke
    0.59
     Coating
    0.59
     marina
    0.57
    感
    0.57
    Steel
    0.57
     STEEL
    0.57
    POSITIVE LOGITS
     crystals
    1.82
     crystal
    1.77
     Crystals
    1.69
    crystals
    1.63
     crystall
    1.63
    crystal
    1.63
     crystallography
    1.58
    Crystal
    1.52
    結晶
    1.51
    晶
    1.50
    Activations Density 0.177%

    No Known Activations