© 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. 11-RES-JB
    11. 22203
    Prev
    Next
    INDEX
    Explanations

    words related to surprises or unexpected events

    oai_token-act-pair · gpt-3.5-turbo

    expressions of shock and surprise

    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
    •layer 11 - surprise

    No Comments

    Negative Logits
    semble
    -0.79
    haar
    -0.71
    rongh
    -0.66
    Rated
    -0.65
     livest
    -0.64
    along
    -0.62
    resp
    -0.62
    nos
    -0.62
    reau
    -0.60
    umbn
    -0.60
    POSITIVE LOGITS
     how
    1.26
     that
    1.12
     why
    1.01
    that
    0.92
     what
    0.91
     seeing
    0.88
     coincidence
    0.84
    why
    0.81
     considering
    0.80
     THAT
    0.80
    Activations Density 0.093%

    No Known Activations