© 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-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 75057
    Prev
    Next
    INDEX
    Explanations

    blog posts, articles, errors

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

    The neuron fires on text prompting or confirming the selection or acceptance of a “best answer” (e.g. instructions to choose or vote an answer as the best).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ویتامین
    -1.38
     چربی
    -1.35
    ׇ
    -1.25
     növény
    -1.19
     دیجیتال
    -1.16
     десер
    -1.15
     måtte
    -1.15
     коронави
    -1.14
     gyóg
    -1.14
     diterima
    -1.13
    POSITIVE LOGITS
     solution
    1.21
    וּ
    1.10
    めます
    1.07
    背影
    1.05
    摞
    1.01
    胸口
    0.98
    וֹ
    0.98
     in
    0.98
    Desen
    0.98
    ּוֹ
    0.97
    Activations Density 0.008%

    No Known Activations