© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8-RES-JB
    11. 13037
    Prev
    Next
    INDEX
    Explanations

    legal and political terms related to laws, regulations, and government actions

    oai_token-act-pair · gpt-3.5-turbo

    references to counts or measures related to political or social actions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     .)
    -0.75
    .)
    -0.65
    JUST
    -0.62
    SPONSORED
    -0.61
    Democratic
    -0.61
     Marin
    -0.61
     Whedon
    -0.60
     Gerr
    -0.59
     Rhode
    -0.59
     Vermont
    -0.58
    POSITIVE LOGITS
     etc
    0.97
     srf
    0.82
     assorted
    0.73
     carbohyd
    0.71
     newcom
    0.70
    etc
    0.70
     pione
    0.69
     captcha
    0.63
    cellaneous
    0.63
     tradem
    0.63
    Activations Density 0.476%

    No Known Activations