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

    Neuronpedia

    Jacobian LensNEW
    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. 2042
    Prev
    Next
    INDEX
    Explanations

    words related to sickness or being unwell

    oai_token-act-pair · gpt-3.5-turbo

    references to sickness or illness

    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
     unlaw
    -0.74
    adr
    -0.66
    ETHOD
    -0.65
     Unch
    -0.65
     compr
    -0.64
    AUT
    -0.63
    successful
    -0.63
    å°Ĩ
    -0.62
    RFC
    -0.61
    wcsstore
    -0.61
    POSITIVE LOGITS
    ening
    1.22
    bay
    1.11
    ened
    1.08
    nesses
    1.01
    estro
    0.91
    ly
    0.89
    ness
    0.89
    est
    0.85
    le
    0.82
    arus
    0.79
    Activations Density 0.011%

    No Known Activations