© 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
    2. Gemma-2-2B
    3. 3-GEMMASCOPE-TRANSCODER-16K
    4. 7229
    Prev
    Next
    INDEX
    Explanations

    a grab bag of seemingly unrelated items: words relating to affection, the word "estate," the fragment "stru," the word "super," sequences like "xe" followed by numbers, and the word "go."

    oai_token-act-pair · gemini-2.0-flash

    loved

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_3/width_16k/average_l0_54
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.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
     Theſe
    -2.03
     myſelf
    -1.98
     Efq
    -1.94
     Jefus
    -1.84
     Anſ
    -1.84
     Monfieur
    -1.82
     Majefty
    -1.80
     itſelf
    -1.76
     Houſe
    -1.76
     ―――――
    -1.76
    POSITIVE LOGITS
     (
    1.11
     I
    1.05
     
    1.03
    <eos>
    0.97
     A
    0.93
     [
    0.92
     J
    0.91
     -
    0.88
     "
    0.87
     C
    0.86
    Activations Density 0.974%

    No Known Activations