© 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. 4-RES-JB
    11. 2483
    Prev
    Next
    INDEX
    Explanations

    phrases containing the word "wit."

    oai_token-act-pair · gpt-3.5-turbo

    instances of wit and cleverness in the text

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

    No Comments

    Negative Logits
    ŃĶ
    -0.63
     tradem
    -0.61
     obligated
    -0.61
     predatory
    -0.60
    Ü
    -0.59
    ysis
    -0.58
     stressful
    -0.57
     Shuttle
    -0.57
     lobbying
    -0.57
    Į
    -0.56
    POSITIVE LOGITS
    hered
    1.31
    chery
    1.11
    sonian
    0.86
    glers
    0.86
    ener
    0.83
    nesses
    0.81
    combe
    0.81
    bol
    0.81
     wit
    0.81
    ician
    0.78
    Activations Density 0.011%

    No Known Activations