© 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. 70896
    Prev
    Next
    INDEX
    Explanations

    apostle, declan, common era

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

    The neuron strongly activates on high-fantasy “world-building” exposition phrases (e.g. “In a world…,” “In the world of…,” descriptions of powers or realms).

    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
    -1.66
     BOOL
    -1.20
    hæng
    -1.17
    וּ
    -1.05
    וֹ
    -1.05
    ־ה
    -1.02
     its
    -1.02
    ֶּ
    -1.02
    ַּ
    -0.99
    AMI
    -0.97
    POSITIVE LOGITS
    ׇ
    1.04
    kram
    0.94
     noirs
    0.94
     hôtels
    0.92
    ὸν
    0.90
    不同
    0.90
     mimpi
    0.89
     other
    0.89
     vodi
    0.89
    rasp
    0.88
    Activations Density 0.007%

    No Known Activations