© 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. 10107
    Prev
    Next
    INDEX
    Explanations

    words and phrases from technical documents for describing processes for producing materials

    oai_token-act-pair · gemini-2.0-flash

    metals, ceramics

    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
    PointerException
    -0.54
     surla
    -0.52
    cency
    -0.52
    gdx
    -0.51
    EndInit
    -0.50
    CloseOperation
    -0.48
    IRM
    -0.48
    ukat
    -0.47
    FormTagHelper
    -0.47
    värr
    -0.47
    POSITIVE LOGITS
     beginnetje
    0.64
    setVerticalGroup
    0.63
     ModelExpression
    0.61
    GEBURTSDATUM
    0.59
    DllImport
    0.54
    ThroughAttribute
    0.52
    vri
    0.52
     يلي
    0.51
     createSlice
    0.50
     Chwiliwch
    0.49
    Activations Density 4.059%

    No Known Activations