© 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. 24-GEMMASCOPE-2-RES-16K
    4. 6837
    Prev
    Next
    INDEX
    Explanations

    replace placeholder text

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

    The neuron fires on abstract, formal nouns or nominalizations that name ideas or concepts (e.g. thought, consideration, attention, advantage).

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

    discourse markers and transition words that introduce suggestions, considerations, or shifts in focus within formal or instructional text.

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

    statements that express cognition or metacognition—noticing, considering, giving attention, acknowledging, or explicitly pointing out information or issues.

    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
    ﻟ
    0.93
    らず
    0.73
    testers
    0.72
     vidéos
    0.71
     کلی
    0.71
     тока
    0.70
    ه‌های
    0.69
     قالت
    0.69
    കളെ
    0.68
    信任
    0.68
    POSITIVE LOGITS
     annually
    0.94
    这种情况
    0.88
     lately
    0.86
     periodically
    0.85
     regularly
    0.78
    ட்டிக்
    0.78
     someday
    0.77
     heretofore
    0.77
     살펴보도록
    0.75
     nowadays
    0.75
    Activations Density 1.054%

    No Known Activations