© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 10841
    Prev
    Next
    INDEX
    Explanations

    phrases containing "sit", "set", or "tuck", often followed by prepositions or adverbs

    oai_token-act-pair · gemini-2.0-flash

    sitting down

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
    ForKey
    -0.60
     favoured
    -0.56
    isRequired
    -0.56
     favored
    -0.52
     jerky
    -0.52
     beroep
    -0.51
     favours
    -0.51
    Portale
    -0.51
     AssemblyVersion
    -0.49
     typedef
    -0.49
    POSITIVE LOGITS
    mybatisplus
    0.95
    +#+
    0.71
     sitting
    0.70
    makeConstraints
    0.70
     ComVisible
    0.70
     duduk
    0.69
    CppMethod
    0.68
     otur
    0.66
    Sitting
    0.63
     متعلقه
    0.62
    Activations Density 1.009%

    No Known Activations