© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 15872
    Prev
    Next
    INDEX
    Explanations

    greater than

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

    The neuron fires strongly on comparative adjectives (especially those ending in “-er,” like Greater, Lesser, higher).

    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
    具合
    0.58
     utc
    0.55
    NESS
    0.54
    Lieutenant
    0.54
    ⃙
    0.54
     alrededor
    0.53
    staticfiles
    0.53
    _^
    0.53
    सनीय
    0.52
    Responses
    0.52
    POSITIVE LOGITS
    than
    1.25
     than
    1.22
     Than
    1.02
     än
    1.00
    Than
    0.96
     smarter
    0.92
     Bigger
    0.87
     kuin
    0.86
     faster
    0.85
     niż
    0.84
    Activations Density 0.216%

    No Known Activations