© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11-RES-JB
    11. 22541
    Prev
    Next
    INDEX
    Explanations

    content related to societal issues, research findings, and public opinions

    oai_token-act-pair · gpt-3.5-turbo

    discussions surrounding opinions and observations related to entertainment, particularly in Hollywood

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ovsky
    -0.65
    quart
    -0.65
    imeter
    -0.63
    endix
    -0.60
    ircraft
    -0.60
    abwe
    -0.60
    Except
    -0.60
    ouver
    -0.59
    Manufact
    -0.59
    ilver
    -0.58
    POSITIVE LOGITS
     homophobia
    1.22
     homophobic
    1.13
     bullying
    1.08
     sexuality
    1.04
     masculinity
    1.03
     LGBT
    1.02
     bisexual
    1.01
     sexism
    1.00
     misogyny
    0.99
     stigma
    0.97
    Activations Density 0.824%

    No Known Activations