© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 10904
    Prev
    Next
    INDEX
    Explanations

    words or phrases that indicate conflict, tension, or significant events, sometimes with an emphasis on feelings or causes

    oai_token-act-pair · gemini-2.0-flash

    stimulation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    клопе
    -0.58
    Personensuche
    -0.53
    pleft
    -0.52
     TRES
    -0.50
     internationaux
    -0.50
    Bata
    -0.48
     AssemblyCulture
    -0.47
     élector
    -0.47
     compé
    -0.47
    ี่ยว
    -0.47
    POSITIVE LOGITS
     Stim
    1.16
     spur
    1.14
     stimulate
    1.13
     spark
    1.13
     stimulation
    1.09
    stim
    1.09
     stimulating
    1.08
     stimulated
    1.05
    Stim
    1.04
     trigger
    1.02
    Activations Density 3.397%

    No Known Activations