Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 8-GEMMASCOPE-TRANSCODER-16K
    4. 698
    Prev
    Next
    INDEX
    Explanations

    This neuron has high activation for what appears to be obfuscated strings, possibly base64 encoded data in HTML, and also activates for text related to scientific publishing such as authorship, competing interests, consent to publish, and ethics approval.

    oai_token-act-pair · gemini-2.0-flash

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.8.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
    posedge
    -0.75
    Hentet
    -0.71
    lccn
    -0.64
    NewUrlParser
    -0.63
     kasarigan
    -0.61
     ComVisible
    -0.59
    titleMargin
    -0.59
    StoryboardSegue
    -0.58
    Predecesor
    -0.58
     pinulongan
    -0.56
    POSITIVE LOGITS
    ABLO
    0.52
     regina
    0.48
     Mediat
    0.47
    ViewFeatures
    0.47
    blower
    0.46
    onView
    0.45
     Insol
    0.45
     zaragoza
    0.45
    泣
    0.44
    ắn
    0.44
    Activations Density 1.220%

    No Known Activations