© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 12-GEMMASCOPE-TRANSCODER-16K
    7. 13035
    Prev
    Next
    INDEX
    Explanations

    multiple things with high activation. Specifically, it activates on comments in code, historical locations, the word "virus", and physical actions. Because of this, it detects passages that are mixed or confusing

    oai_token-act-pair · gemini-2.0-flash

    code snippets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.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
     imported
    -0.68
     arrival
    -0.66
    imported
    -0.64
     arrivée
    -0.64
     arriving
    -0.63
    arrival
    -0.60
     Arrival
    -0.59
     Arrive
    -0.59
     arriv
    -0.57
     Tourists
    -0.57
    POSITIVE LOGITS
     AssemblyProduct
    0.69
     للمعارف
    0.64
    Autoritní
    0.63
    Hentet
    0.57
     <<<<<<<<<<<<<<
    0.57
     EconPapers
    0.56
    سطس
    0.56
    contentLoaded
    0.53
    memoized
    0.52
    SourceChecksum
    0.50
    Activations Density 2.025%

    No Known Activations