© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14707
    Prev
    Next
    INDEX
    Explanations

    comparisons, especially using words like "lower", "higher", and "greater", and also finds words related to separate instances or parts of things

    oai_token-act-pair · gemini-2.0-flash

    comparative measurements or values

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

    Comparing low/high

    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
    AsUp
    -1.09
     GenerationType
    -1.01
     doubtnut
    -0.94
    ſelf
    -0.91
     ſind
    -0.90
     disambiguazione
    -0.88
    ſelves
    -0.88
     تضيفلها
    -0.85
     ſtate
    -0.85
     ſy
    -0.85
    POSITIVE LOGITS
    .
    0.46
     [
    0.46
     (
    0.45
    ,
    0.43
     —
    0.41
     in
    0.39
     dibanding
    0.39
     under
    0.39
     hậu
    0.38
    EXTERN
    0.38
    Activations Density 3.785%

    No Known Activations