© 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. 1951
    Prev
    Next
    INDEX
    Explanations

    words related to environmental remediation of contaminated soil

    oai_token-act-pair · gemini-2.0-flash

    terms related to waste management or environmental issues

    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
     Monfieur
    -0.58
     houſe
    -0.56
     raiſ
    -0.55
    Παραπομπές
    -0.52
     itſelf
    -0.51
     ſever
    -0.50
     Caixa
    -0.50
     megh
    -0.50
     ListTile
    -0.49
     Perſ
    -0.49
    POSITIVE LOGITS
    TagMode
    0.78
     Super
    0.60
     dio
    0.57
    dio
    0.54
     super
    0.54
    Super
    0.53
    Jeografia
    0.52
    featureID
    0.52
     spent
    0.52
     CER
    0.52
    Activations Density 2.674%

    No Known Activations