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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 4819
    Prev
    Next
    INDEX
    Explanations

    mentions of mathematical and physics concepts, theorems, and people

    oai_token-act-pair · gemini-2.0-flash

    dash or hyphen-like characters

    np_token-act-pair-logits · gpt-4o-mini

    mathematical/physical theorems

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
     faſt
    -0.90
     purpoſe
    -0.86
     Theſe
    -0.85
     uſed
    -0.83
     leaſt
    -0.81
     Monfieur
    -0.81
     myſelf
    -0.77
     houſe
    -0.75
     pleaſure
    -0.75
     Cæsar
    -0.75
    POSITIVE LOGITS
    –
    0.63
    —
    0.58
    ‒
    0.57
    --
    0.57
    ̍t
    0.57
    ―
    0.54
    SFD
    0.54
     —
    0.52
     est
    0.52
     --
    0.50
    Activations Density 2.265%

    No Known Activations