© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 12923
    Prev
    Next
    INDEX
    Explanations

    (Part 1) words/phrases that signify a change in something over time.

    oai_token-act-pair · gemini-2.0-flash

    decrease/decline

    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
     betweenstory
    -0.50
    uffy
    -0.49
    hoop
    -0.48
     calmed
    -0.47
     Rado
    -0.47
    љено
    -0.46
    ISupport
    -0.46
    ριά
    -0.46
    Gt
    -0.45
    gop
    -0.45
    POSITIVE LOGITS
    
    0.59
    Клуб
    0.58
     protoimpl
    0.58
     Chwiliwch
    0.57
    Hentet
    0.54
    Havolalar
    0.54
    -------
    0.52
     poveznice
    0.51
    Naissance
    0.50
     w
    0.49
    Activations Density 1.348%

    No Known Activations