© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 4360
    Prev
    Next
    INDEX
    Explanations

    the word "box", especially when capitalized

    oai_token-act-pair · gemini-2.0-flash

    instances of the word 'box'

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

    box

    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
     Roskov
    -0.62
    AddTagHelper
    -0.60
    OGND
    -0.55
    findpost
    -0.53
     EdgeInsets
    -0.52
     hObject
    -0.51
    AndroidJUnit
    -0.50
    CloseOperation
    -0.48
    DrawerToggle
    -0.46
    RTLU
    -0.46
    POSITIVE LOGITS
     שוליים
    0.69
    ClientRect
    0.61
    வு
    0.54
    ձ
    0.54
    */].
    0.53
    Identyfik
    0.53
     Résultats
    0.52
     חיצוניים
    0.52
    MLLoader
    0.51
    mme
    0.51
    Activations Density 2.149%

    No Known Activations