© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 4935
    Prev
    Next
    INDEX
    Explanations

    words related to satire, humor, sarcasm, and irony

    oai_token-act-pair · gpt-3.5-turbo

    elements of satire and humor in discussions or critiques

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sacrific
    -0.73
    enium
    -0.63
     Horizons
    -0.61
    conservancy
    -0.60
    ignty
    -0.60
    surv
    -0.59
    )",
    -0.58
     determines
    -0.58
    "},"
    -0.58
     bottleneck
    -0.58
    POSITIVE LOGITS
     rhetorical
    1.10
     mockery
    0.95
     journalistic
    0.91
     chuckle
    0.90
     understatement
    0.88
     disingen
    0.83
     disson
    0.82
     tone
    0.82
     irony
    0.81
     jarring
    0.80
    Activations Density 1.154%

    No Known Activations