© 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. 4-RES-JB
    11. 7447
    Prev
    Next
    INDEX
    Explanations

    negative actions or attributes associated with dehumanizing, demonizing, stigmatizing, or vilifying individuals or groups

    oai_token-act-pair · gpt-3.5-turbo

    words related to social stigmatization and dehumanization

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

    No Comments

    Negative Logits
    UTERS
    -0.69
    oret
    -0.62
    negie
    -0.60
     enthusi
    -0.59
    uid
    -0.59
    INC
    -0.58
    cffff
    -0.57
     Prospect
    -0.57
    stead
    -0.57
    NH
    -0.56
    POSITIVE LOGITS
     slurs
    0.92
    imaru
    0.90
     stereotypes
    0.88
     stigma
    0.80
     prejudice
    0.77
     vil
    0.77
     insults
    0.76
     bullies
    0.76
     stigmat
    0.75
     dehuman
    0.74
    Activations Density 0.079%

    No Known Activations