© 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. 0-RES-JB
    11. 22488
    Prev
    Next
    INDEX
    Explanations

    strings of characters that serve as separators or decorative elements within a text

    oai_token-act-pair · gpt-3.5-turbo

    repetitive or structured text elements, possibly indicating formatting or layout features in the document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    xs
    -0.77
     Saban
    -0.73
     cascade
    -0.72
    andre
    -0.70
    ramid
    -0.68
     Mn
    -0.68
     UA
    -0.67
     Atl
    -0.66
     outp
    -0.66
     Mex
    -0.66
    POSITIVE LOGITS
    ______
    2.70
    _______
    2.56
    _____
    2.06
    ________
    1.80
     ______
    1.63
    ––
    1.60
    ——
    1.47
    ____
    1.38
    ________________
    1.26
     Chester
    1.17
    Activations Density 0.055%

    No Known Activations