© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 6431
    Prev
    Next
    INDEX
    Explanations

    words and phrases that make an argument

    oai_token-act-pair · gemini-2.0-flash

    technical terms and classifications

    np_token-act-pair-logits · gpt-4o-mini

    finance/business/economics

    np_max-act-logits · gemini-2.0-flash
    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
    yaitu
    -0.54
     hâte
    -0.53
    اریخ
    -0.52
    Regarding
    -0.51
    expandindo
    -0.50
    monių
    -0.49
     Regarding
    -0.48
     aprobado
    -0.47
     profondità
    -0.47
    stateProvider
    -0.46
    POSITIVE LOGITS
     severally
    0.72
     Okada
    0.65
    IMPORTED
    0.64
     sensitization
    0.62
     artistes
    0.60
     оригіналу
    0.60
     enumeration
    0.59
     تانيه
    0.57
     correctes
    0.57
     computation
    0.56
    Activations Density 16.778%

    No Known Activations