© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 13253
    Prev
    Next
    INDEX
    Explanations

    information related to online content creation, such as websites, official statements, and content formats like videos and prototypes

    oai_token-act-pair · gpt-3.5-turbo

    content related to promotional claims and product descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĪĴ
    -0.76
    iencies
    -0.75
     satell
    -0.72
    chwitz
    -0.72
     continuity
    -0.72
     restructuring
    -0.70
     renegoti
    -0.70
     involuntary
    -0.69
    ©¶æ
    -0.68
     consequential
    -0.67
    POSITIVE LOGITS
    Redditor
    1.05
     Featuring
    1.03
     Users
    0.96
    âĸ
    0.95
    via
    0.94
    Scroll
    0.94
     Yep
    0.94
     Scroll
    0.93
    Users
    0.93
     ®
    0.90
    Activations Density 0.560%

    No Known Activations