© 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. 10854
    Prev
    Next
    INDEX
    Explanations

    filler words and phrases whose purpose is to add emphasis or make a statement less bold

    oai_token-act-pair · gemini-2.0-flash

    words emphasizing inclusivity or universality

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

    ever, every

    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
    None
    -0.51
    none
    -0.46
     None
    -0.44
    
    -0.41
    TestingModule
    -0.41
    nés
    -0.41
    Parcelize
    -0.40
     wir
    -0.40
     миру
    -0.40
    autres
    -0.40
    POSITIVE LOGITS
     ever
    1.28
     every
    1.09
     evermore
    1.05
     Ever
    0.91
     EVER
    0.88
     EVERY
    0.88
    Ever
    0.88
     ev
    0.87
    EVERY
    0.85
     absolutely
    0.84
    Activations Density 0.737%

    No Known Activations