© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 7041
    Prev
    Next
    INDEX
    Explanations

    HTML elements related to images in a webpage

    oai_token-act-pair · gpt-3.5-turbo

    image and media-related tags and elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ese
    -0.87
    kamp
    -0.80
    ebus
    -0.78
    Ĥ¬
    -0.73
    etheless
    -0.73
    ŃĶ
    -0.72
     Griff
    -0.71
    theless
    -0.70
    hower
    -0.68
    ĨĴ
    -0.68
    POSITIVE LOGITS
    ãĤ¨ãĥ«
    1.06
    img
    1.05
    uploads
    0.93
    wcsstore
    0.91
     src
    0.88
    Upload
    0.85
     img
    0.82
    thumbnails
    0.81
    ãĤ´
    0.81
    Images
    0.79
    Activations Density 0.010%

    No Known Activations