© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15217
    Prev
    Next
    INDEX
    Explanations

    mentions of offices and locations and their relation to companies

    oai_token-act-pair · gemini-2.0-flash

    say "office"

    np_max-act-logits · gemini-2.0-flash

    variations of the word "office"

    np_token-act-pair-logits · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
     forKey
    -0.45
     élevées
    -0.41
    ALARM
    -0.38
     Chất
    -0.37
     piłkarz
    -0.37
    MessageTagHelper
    -0.36
    Autores
    -0.36
    CloseOperation
    -0.36
     abon
    -0.36
     nguyễn
    -0.36
    POSITIVE LOGITS
     office
    2.28
    office
    1.97
     offices
    1.85
     Office
    1.81
    Office
    1.74
     OFFICE
    1.74
    OFFICE
    1.60
    offices
    1.52
     Offices
    1.52
     Büro
    1.48
    Activations Density 1.210%

    No Known Activations