© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 666
    Prev
    Next
    INDEX
    Explanations

    words related to being surprised

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "surprise" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 1 - surprise

    No Comments

    Negative Logits
    ©¶æ
    -1.07
    nan
    -0.87
    urgy
    -0.77
    ergy
    -0.73
    ccording
    -0.72
    İĭ
    -0.72
    Ķ
    -0.71
    lim
    -0.68
    oran
    -0.67
    uties
    -0.67
    POSITIVE LOGITS
     surprise
    1.17
     Surprise
    1.06
     surprises
    1.05
    ingly
    0.88
     darling
    0.85
     surpr
    0.81
     onlook
    0.75
     aloud
    0.72
     aston
    0.69
    aneously
    0.69
    Activations Density 0.011%

    No Known Activations