© 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. 1-RES-JB
    11. 661
    Prev
    Next
    INDEX
    Explanations

    references to illegal or forbidden activities, especially pertaining to drugs or trade

    oai_token-act-pair · gpt-3.5-turbo

    references to illegal activities and substances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    soType
    -0.81
    packs
    -0.78
    frames
    -0.75
    Point
    -0.73
    ILA
    -0.71
    SAY
    -0.70
    ds
    -0.68
    onement
    -0.68
    anders
    -0.67
    iph
    -0.67
    POSITIVE LOGITS
     illicit
    1.10
     lic
    0.94
     trafficking
    0.86
     smuggling
    0.84
     substances
    0.83
     narcotics
    0.80
     clandestine
    0.77
     overfl
    0.77
     drugs
    0.75
     stash
    0.75
    Activations Density 0.011%

    No Known Activations