© 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. 14878
    Prev
    Next
    INDEX
    Explanations

    apostrophes preceded by a letter, which is an artifacting of tokenization around apostrophes, and may or may not indicate possessives or contractions.

    oai_token-act-pair · gemini-2.0-flash

    apostrophe

    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
    ,
    -1.94
     (
    -1.92
     “
    -1.84
     ‘
    -1.76
     
    -1.75
    -
    -1.69
    /
    -1.64
    :
    -1.59
    <eos>
    -1.55
     in
    -1.52
    POSITIVE LOGITS
     myſelf
    3.73
     Efq
    3.48
     itſelf
    3.47
     Monfieur
    3.34
     Theſe
    3.20
     pleaſure
    3.09
     purpoſe
    3.03
     Jefus
    3.03
     themſelves
    2.98
     houſe
    2.98
    Activations Density 0.737%

    No Known Activations