© 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-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 4489
    Prev
    Next
    INDEX
    Explanations

    French words and phrases alongside fragments of URLs and code, indicative of a technical conversation with mixed languages.

    oai_token-act-pair · gemini-2.0-flash

    multi-lingual text

    np_max-act-logits · gemini-2.0-flash

    French

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     son
    -0.85
    /**
    -0.85
    Son
    -0.78
     Son
    -0.71
     sur
    -0.66
     sa
    -0.66
    DockStyle
    -0.66
    <?
    -0.66
     незавершена
    -0.66
     Оно
    -0.65
    POSITIVE LOGITS
     her
    0.93
     his
    0.69
     their
    0.54
    writeFieldEnd
    0.53
     AssemblyCompany
    0.52
     utafitiHapana
    0.48
    their
    0.47
    NOPQRST
    0.47
    twimg
    0.47
     HER
    0.45
    Activations Density 0.791%

    No Known Activations