© 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. 24-GEMMASCOPE-2-RES-16K
    7. 12366
    Prev
    Next
    INDEX
    Explanations

    descriptive terms followed by related nouns

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

    The neuron fires on concrete material or color terms (e.g. “silver,” “gold,” “leather,” “chocolate”) used in product descriptions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    words describing materials or physical properties, particularly metals and textures.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    descriptive terms for tangible items—materials, shapes, and technical specs—common in product or instructional text.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_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
    向けの
    0.62
    一名
    0.56
     CPUs
    0.55
     வகையான
    0.55
     girlfriends
    0.55
     midfielders
    0.53
     nonprofits
    0.52
     submarines
    0.52
     millennials
    0.52
     scholarships
    0.52
    POSITIVE LOGITS
     aforesaid
    0.57
     tersebut
    0.56
     purport
    0.54
    当該
    0.52
     해당
    0.49
     nêu
    0.46
     tient
    0.46
     doet
    0.45
     ensuing
    0.45
    \}.
    0.45
    Activations Density 0.007%

    No Known Activations