© 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. 5790
    Prev
    Next
    INDEX
    Explanations

    the concept of surprise or lack thereof

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to surprising assertions

    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
    rongh
    -0.78
    \\\\\\\\\\\\\\\\
    -0.67
    ////////////////
    -0.63
     embr
    -0.62
    qus
    -0.61
    construct
    -0.61
    umbn
    -0.61
     exting
    -0.61
     raft
    -0.60
    ////////////////////////////////
    -0.60
    POSITIVE LOGITS
     then
    0.90
     that
    0.78
     why
    0.75
     considering
    0.73
     given
    0.64
    that
    0.64
    imaru
    0.61
    Reviewer
    0.61
    ²¾
    0.60
     whatsoever
    0.60
    Activations Density 0.090%

    No Known Activations