© 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. 11009
    Prev
    Next
    INDEX
    Explanations

    multilingual academic and biblical text

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

    The neuron strongly fires on tokens characteristic of Biblical or archaic scripture passages—e.g. verse numbers in Roman numerals and archaic “And I sawe/ascende”‐style wording.

    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
    ּוֹ
    -1.69
    tersebut
    -1.58
    Toeschouwers
    -1.56
    וּ
    -1.56
    ִּ
    -1.52
    ־ה
    -1.48
     dirilis
    -1.48
    Jangan
    -1.48
    ַּ
    -1.42
     бампер
    -1.41
    POSITIVE LOGITS
    ὴν
    1.58
     hefyd
    1.55
     [
    1.48
     francès
    1.38
    >[
    1.37
    ">[
    1.37
    地看着
    1.34
     αὐτ
    1.34
    itting
    1.32
    []){
    1.31
    Activations Density 0.005%

    No Known Activations