© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33325
    Prev
    Next
    INDEX
    Explanations

    DIY construction and creation

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

    The neuron fires on the procedural, step‐by‐step “DIY” or how-to instruction language (e.g. verbs like made, gluing, built, by, into, 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
    ְּ
    -1.52
    ִּ
    -1.51
    ּוֹ
    -1.49
    ַּ
    -1.45
     бампер
    -1.38
    ֶּ
    -1.28
     أيض
    -1.18
     ktorá
    -1.17
    Irlande
    -1.17
     सत्यापित
    -1.16
    POSITIVE LOGITS
     homemade
    1.54
     ideas
    1.45
     χρησιμοποι
    1.41
    استفاده
    1.37
     DIY
    1.29
    ׇ
    1.28
     استفاده
    1.28
     construction
    1.27
     design
    1.27
     creation
    1.25
    Activations Density 0.066%

    No Known Activations