© 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 EvalsExportsinterp-engineNEWAPI 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. 12479
    Prev
    Next
    INDEX
    Explanations

    words related to leaks or leaking information

    oai_token-act-pair · gpt-3.5-turbo

    references to leaks of information or documents

    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
    iterranean
    -0.80
    stad
    -0.71
    igel
    -0.63
     Dragons
    -0.63
    aldi
    -0.62
    aram
    -0.62
    anu
    -0.62
     Malays
    -0.62
    lication
    -0.61
    GRE
    -0.60
    POSITIVE LOGITS
    acious
    0.90
     leaks
    0.88
     leak
    0.85
     leaked
    0.84
     leaking
    0.83
    Leaks
    0.77
     confidential
    0.76
     secrets
    0.75
    ster
    0.75
    ariat
    0.73
    Activations Density 0.034%

    No Known Activations