© 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. 13092
    Prev
    Next
    INDEX
    Explanations

    the words "down" and "up", often in combination

    oai_token-act-pair · gemini-2.0-flash

    down

    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
     down
    -1.58
    down
    -1.25
     up
    -1.05
     DOWN
    -1.05
     Down
    -1.03
    Down
    -0.96
    DOWN
    -0.79
     Up
    -0.79
    up
    -0.72
    Up
    -0.69
    POSITIVE LOGITS
    Hochspringen
    0.66
     Secondo
    0.54
     Versuche
    0.52
     samband
    0.51
     verschieden
    0.51
     oculto
    0.50
     Giappone
    0.50
     secreta
    0.50
     Noviembre
    0.50
     unterschiedlich
    0.50
    Activations Density 0.131%

    No Known Activations