© 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. 1-RES-JB
    11. 21910
    Prev
    Next
    INDEX
    Explanations

    keywords related to violence, crime, and conflict

    oai_token-act-pair · gpt-3.5-turbo

    references to criminals or violent individuals, specifically labeled as "thugs."

    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
    igm
    -0.78
    ced
    -0.76
    itionally
    -0.73
    rontal
    -0.73
    elaide
    -0.72
    venue
    -0.71
    USE
    -0.67
    EP
    -0.67
    mary
    -0.67
    fman
    -0.66
    POSITIVE LOGITS
     thugs
    1.14
     thug
    1.08
    gery
    0.90
    ocrats
    0.78
     mobs
    0.76
    oby
    0.75
     Droid
    0.74
    atics
    0.72
    sie
    0.69
     Thieves
    0.69
    Activations Density 0.016%

    No Known Activations