© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 10401
    Prev
    Next
    INDEX
    Explanations

    LaTeX mathematical formulas, especially those involving partial derivatives

    oai_token-act-pair · gemini-2.0-flash

    partial

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     Monfieur
    -2.34
     Efq
    -2.34
     itſelf
    -2.33
     myſelf
    -2.31
     pleaſure
    -2.06
     متعلقه
    -2.06
     ſeveral
    -2.03
     للاسماء
    -2.03
     becauſe
    -2.02
     themſelves
    -2.02
    POSITIVE LOGITS
     
    1.49
    ,
    1.13
     (
    1.11
    <eos>
    1.05
    ↵↵
    1.05
    ↵
    1.02
     A
    1.00
    .
    0.99
     in
    0.96
     “
    0.95
    Activations Density 15.255%

    No Known Activations