© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 10612
    Prev
    Next
    INDEX
    Explanations

    the word "understood" and words related to it, such as inventions and responsibility

    oai_token-act-pair · gemini-2.0-flash

    understood

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     Ministero
    -0.78
     comigo
    -0.78
     comprendre
    -0.76
     consciência
    -0.76
     understands
    -0.76
     understood
    -0.75
     noires
    -0.75
     deschis
    -0.75
     Zwiebel
    -0.75
     retrô
    -0.74
    POSITIVE LOGITS
    ")));
    0.72
    rawn
    0.67
    ()")
    0.65
    ]),
    0.65
    ()):
    0.65
    )");
    0.65
    \)
    0.64
    ¯¯
    0.63
    ingue
    0.63
    dians
    0.63
    Activations Density 1.166%

    No Known Activations