© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67284
    Prev
    Next
    INDEX
    Explanations

    quiz challenge questions activities

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

    The neuron strongly activates on very common, high‐frequency function words (short articles, pronouns, infinitival “to,” simple verbs, etc.).

    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
     zwi
    -1.10
    氅
    -1.08
     الحكوم
    -1.08
     bezw
    -1.07
     Lunes
    -1.06
    actrice
    -1.02
     الحكومة
    -1.01
     ストラップ
    -0.99
     ekster
    -0.99
     biografi
    -0.98
    POSITIVE LOGITS
     quiz
    1.34
     challenge
    1.32
     questions
    1.32
     food
    1.30
     activities
    1.30
    ônus
    1.20
     activity
    1.17
     foods
    1.16
     quizzes
    1.15
     question
    1.14
    Activations Density 0.068%

    No Known Activations