© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 12064
    Prev
    Next
    INDEX
    Explanations

    a mix of capitalized words, acronyms, and terms related to study design and location.

    oai_token-act-pair · gemini-2.0-flash

    ESG

    np_max-act · gemini-2.0-flash

    say "space" Method used: 2 — reason: tokens often start with a leading space — say "space"

    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_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
    RegressionTest
    -0.76
     الحره
    -0.68
     kasarigan
    -0.58
    \}\\
    -0.56
    hueta
    -0.50
     gug
    -0.48
    itag
    -0.47
    SHO
    -0.47
    ])){
    -0.46
    Activité
    -0.46
    POSITIVE LOGITS
    <bos>
    0.80
     StatefulWidget
    0.64
    BorderLayout
    0.61
    tagHelperRunner
    0.57
    ActionCreators
    0.56
     doInBackground
    0.55
     PhpStorm
    0.53
     calloc
    0.50
    bootstrapcdn
    0.50
     ElementRef
    0.50
    Activations Density 17.118%

    No Known Activations