© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 12293
    Prev
    Next
    INDEX
    Explanations

    a bunch of common words, especially pronouns, prepositions, conjunctions, and the word "David".

    oai_token-act-pair · gemini-2.0-flash

    Goalkeepers and legal contexts

    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
     تضيفلها
    -0.97
    Portail
    -0.90
    ArrowToggle
    -0.77
     raiſ
    -0.77
    脚注の使い方
    -0.75
    FromNib
    -0.73
     purpoſe
    -0.73
    ImageContext
    -0.72
    NewLabel
    -0.72
    İstinadlar
    -0.72
    POSITIVE LOGITS
    is
    0.43
    SC
    0.42
    Con
    0.42
     Con
    0.41
    des
    0.40
    o
    0.40
    UNCH
    0.40
    E
    0.40
    Da
    0.39
     Mat
    0.39
    Activations Density 0.508%

    No Known Activations