© 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. 2-RES-JB
    11. 470
    Prev
    Next
    INDEX
    Explanations

    the acronym 'SWAT'

    oai_token-act-pair · gpt-4-turbo

    references to SWAT teams

    oai_token-act-pair · gpt-3.5-turbo

    mentions of SWAT teams and associated law enforcement terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    living
    -0.71
    birth
    -0.70
    iments
    -0.67
    spring
    -0.66
    pend
    -0.61
    south
    -0.61
    drawn
    -0.61
    prem
    -0.61
     Franken
    -0.61
    reviewed
    -0.61
    POSITIVE LOGITS
     SWAT
    0.97
    ormal
    0.74
    ixel
    0.73
     swat
    0.72
    ilitary
    0.71
     departments
    0.70
     squads
    0.69
    ELY
    0.69
    antha
    0.68
    auld
    0.67
    Activations Density 0.011%

    No Known Activations