© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 20-GEMMASCOPE-TRANSCODER-16K
    7. 10162
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to be finding elements of fantastical stories, especially words associated with kingdoms, magic, and legendary figures

    oai_token-act-pair · gemini-2.0-flash

    invented fantasy world

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    tagHelperRunner
    -0.83
    parsedMessage
    -0.79
    writeFieldEnd
    -0.78
     surla
    -0.70
    Personendaten
    -0.66
     الرياضيه
    -0.65
    BufferException
    -0.64
     disambiguazione
    -0.63
     supérieurs
    -0.61
    Демографія
    -0.61
    POSITIVE LOGITS
    ensa
    0.51
    zension
    0.47
    鑑
    0.47
    wsdl
    0.45
    SPJ
    0.45
    Attrib
    0.44
    ələ
    0.44
     array
    0.44
     date
    0.44
     intent
    0.44
    Activations Density 0.364%

    No Known Activations