© 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. 3-RES-JB
    11. 5406
    Prev
    Next
    INDEX
    Explanations

    phrases related to illegal actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "illegally" and related phrases regarding unlawful activities

    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
    nil
    -0.81
     enthus
    -0.78
    oleon
    -0.78
    ieu
    -0.71
    andals
    -0.70
    ema
    -0.69
    antics
    -0.69
    erer
    -0.68
    eth
    -0.68
    addons
    -0.68
    POSITIVE LOGITS
     illegally
    0.90
     copied
    0.87
     downloaded
    0.81
     detained
    0.78
     infringing
    0.75
     downloading
    0.75
     intercepted
    0.74
     obtained
    0.74
    IFIED
    0.72
     disreg
    0.71
    Activations Density 0.013%

    No Known Activations