© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 511
    Prev
    Next
    INDEX
    Explanations

    that spans, goes, accompanies

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

    This neuron selectively spikes on longer, content‐heavy words (i.e. multi-syllable or “complex” terms) rather than short or function words.

    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
    ที่เป็น
    0.39
     или
    0.38
    или
    0.35
    ที่จะ
    0.35
     மற்றும்
    0.34
     กับ
    0.33
    ())
    0.33
    )+
    0.32
    กับ
    0.32
     หรือ
    0.32
    POSITIVE LOGITS
     underlies
    0.75
     encompasses
    0.70
     transcends
    0.70
     includes
    0.70
     occurs
    0.70
     permeates
    0.70
     characterizes
    0.69
     incorporates
    0.68
     embodies
    0.68
     originates
    0.67
    Activations Density 0.709%

    No Known Activations