© 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. 11-RES-JB
    11. 1358
    Prev
    Next
    INDEX
    Explanations

    words related to societal and political issues, especially negative and critical ones

    oai_token-act-pair · gpt-3.5-turbo

    themes related to societal issues and the impacts of governance

    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
    omething
    -0.70
     wording
    -0.70
    ":[{"
    -0.69
     behaviours
    -0.69
    agraph
    -0.66
    ints
    -0.65
    ENTS
    -0.65
     ABV
    -0.65
    ensibly
    -0.64
     sentiments
    -0.64
    POSITIVE LOGITS
     cess
    1.27
     battleground
    1.21
     wasteland
    1.19
     mess
    1.10
     playground
    1.10
     battlefield
    1.07
     maze
    1.04
     swamp
    0.99
     fortress
    0.99
     graveyard
    0.99
    Activations Density 0.239%

    No Known Activations