© 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. 0-RES-JB
    11. 5284
    Prev
    Next
    INDEX
    Explanations

    references to vulnerable groups like orphans or prostitutes

    oai_token-act-pair · gpt-3.5-turbo

    references to vulnerable groups, specifically orphans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĥĥãĥī
    -0.88
    76561
    -0.79
    cheat
    -0.79
    Rat
    -0.79
    Lot
    -0.75
    rack
    -0.74
    lot
    -0.73
    SpaceEngineers
    -0.72
    soDeliveryDate
    -0.72
    EVA
    -0.71
    POSITIVE LOGITS
     orphans
    0.89
     Department
    0.89
    yll
    0.70
     daily
    0.67
    xual
    0.65
    acters
    0.65
     brig
    0.65
    tymology
    0.64
     militia
    0.63
     academy
    0.63
    Activations Density 0.000%

    No Known Activations