© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 3831
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to activate on a mix of Arabic text, Java import statements, "un" prefixes, and the phrase "For the full"

    oai_token-act-pair · gemini-2.0-flash

    foreign words and characters

    np_token-act-pair-logits · gpt-4o-mini

    code and foreign languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Enllaces
    -0.50
    opsis
    -0.50
    LookAnd
    -0.50
     l
    -0.50
     Nach
    -0.48
     berk
    -0.48
    IndentedString
    -0.47
     F
    -0.47
     Lang
    -0.47
     "
    -0.47
    POSITIVE LOGITS
    Obrázky
    0.66
    มี
    0.62
    ihnachten
    0.61
     Bhd
    0.60
     woodpecker
    0.59
     useAuth
    0.57
     azy
    0.57
    -#
    0.57
    NUMX
    0.56
     Signalez
    0.56
    Activations Density 0.156%

    No Known Activations