© 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. 6-GEMMASCOPE-TRANSCODER-16K
    7. 14884
    Prev
    Next
    INDEX
    Explanations

    words related to unexpectedness or discovery

    oai_token-act-pair · gemini-2.0-flash

    surprise

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_6/width_16k/average_l0_95
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.6.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
     surprise
    -1.38
    surprise
    -1.30
     Surprise
    -1.26
     surprised
    -1.22
     surprises
    -1.20
    Surprise
    -1.15
    surprised
    -1.11
     sorpresa
    -1.09
     surpresa
    -1.01
     Überras
    -1.01
    POSITIVE LOGITS
    værende
    0.53
     share
    0.51
    veloppement
    0.50
    newtheorem
    0.50
     reflectivity
    0.49
     conductive
    0.48
     Khr
    0.48
    sega
    0.48
     complete
    0.48
     Sangu
    0.48
    Activations Density 17.308%

    No Known Activations