© 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. 3-RES-JB
    11. 6360
    Prev
    Next
    INDEX
    Explanations

    words related to sarcastic or snide remarks

    oai_token-act-pair · gpt-3.5-turbo

    words associated with humor, particularly playful or mocking language

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unequal
    -0.72
     absence
    -0.68
     suff
    -0.68
    EVA
    -0.66
     tert
    -0.66
     triangle
    -0.66
    ICAN
    -0.66
     disparate
    -0.66
     concess
    -0.65
     complementary
    -0.65
    POSITIVE LOGITS
    puff
    0.85
    vous
    0.82
    oslav
    0.79
    glers
    0.79
    hound
    0.76
    pants
    0.75
    eps
    0.74
    hawk
    0.74
    yp
    0.74
    tty
    0.73
    Activations Density 0.092%

    No Known Activations